Abstract: An image processing device that sequentially generates a synthetic image by joining a first image, which is a single image or constituted by a plurality of the images joined together, and a second image that is input, every time when the second image is input, includes: an overlapping region obtainment portion 14 that obtains a motion vector on the basis of the second image and an image input immediately before the second image among the images constituting the first image, and obtains an overlapping region where the image input immediately before and the second image overlap each other on the basis of the obtained motion vector; a brightness-value conversion factor calculation portion 16 that calculates a brightness conversion factor for converting a brightness value of the first image or a brightness value of the second image on the basis of the brightness value of the first image and the brightness value of the second image in the overlapping region so that a change in the brightness values between the fir

Abstract: A method and apparatus for partially up/downscaling an image encoded on a macroblock basis utilizing a computer. The method and apparatus performs operations of: storing the encoded image; creating map data from bitstream of the encoded image to decode at least one macroblock of the encoded image, creating a shrunken image of a predetermined size based on a screen resolution of a display device, storing the map data and the shrunken image so as to relate with the encoded image in the storage device; outputting the shrunken image related with the encoded image to be displayed based on a control request received from an input device, determining at least one macroblock to be decoded based on a display area of the shrunken image; partially decoding the encoded image for the determined macroblock using the map data; and outputting to the display device, the image data of the display area of the partially decoded image.

Abstract: An image processing device includes an image input portion, a motion vector acquisition portion and a distortion component estimation portion. The image input portion inputs a target image which is an image to be processed. The motion vector acquisition portion acquires a motion vector generated by relative movement with respect to a subject drawn in the target image when an imaging device which has generated the target image generates the target image. The distortion component estimation portion estimates a rolling shutter distortion component of the target image based on the motion vector.

Abstract: An image processing apparatus 1 includes: a placement information acquisition section 40 which acquires, for each image sensor, placement information of the image sensors with respect to the imaging section 10; a movement information acquisition section 30 which acquires movement information of the imaging device; and an output area acquisition section which acquires, for every image sensor, position information of a second output area within the second frame image which corresponds to a first output area within the first frame image. The output area acquisition section acquires: a second output area A12 of a base image sensor based on position information of a first output area A11 determined by the placement information and based on the movement information; and a second output area A22 of another image sensor based on the placement information of the base image sensor, the placement information of the other image sensor, and the movement information.

Abstract: An image identification method for classifying block images of input image data into one of predetermined categories; the method includes the steps of: dividing image data into multiple blocks to produce block images, processing the feature quantity of each block image by their color space information and frequency component, learning separating hyperplanes that indicate boundaries of each category by reading in training data image that have labeled categories for each block and processing image feature quantity for each block of an training data image, and classifying respective block image to a category according to the distance from the separating hyperplane of each category for a newly acquired image to obtain the image feature quantity of block images. An imaging apparatus implementing the image identification method noted above is also disclosed.

Abstract: An image processing device includes an image input portion, a motion vector acquisition portion and a distortion component estimation portion. The image input portion inputs a plurality of frame images. The motion vector acquisition portion acquires an amount of parallel movement between two frame images. The distortion component estimation portion performs approximation such that a motion matrix indicating a motion between the frame images consists only of a parallel movement component and a rolling shutter distortion component, uses the amount of parallel movement to calculate, for each of the frame images, a tentative distortion coefficient included in the rolling shutter distortion component and uses the plurality of tentative distortion coefficients calculated to estimate a distortion coefficient.

Abstract: The invention provides an image identification device uses a separating plane to classify block images into the categories. The image identification device includes a target image input unit inputting a target image, a block image generation unit generates block images, a feature quantity computing unit computes feature quantities of the block images, and a category determination unit determines whether the block images are classified into the categories or not. The feature quantity computing unit uses local feature quantities of the block images and a global feature quantity of the target image as a whole, and also in a feature quantity space using features of the block images as coordinate axes, uses coordinate positions of feature quantity vectors optional areas in the feature quantity space to count the block images and causes the global feature quantity to include the number of the block images thus counted.

Abstract: The invention provides an image identification device that classifies block images obtained by dividing a target image into predetermined categories, using a separating plane learning of which has been completed in advance for each of the categories. The image identification device includes a target image input unit inputs the target image, a block image generation unit divides the target image into blocks to generate the block images, a feature quantity computing unit computes feature quantities of the block images, and a category determination unit determines whether the block images are classified into one of the categories or not, using the separating plane and coordinate positions corresponding to magnitudes of feature quantities of the block images in a feature quantity space, wherein the feature quantity computing unit uses, as a feature quantity of a given target block image, local feature quantities and a global feature quantity.

Abstract: A moving object detecting device 1 that detects a moving object by using an image includes a motion degree obtaining portion 11 that obtains a motion degree of a pixel between image frames, a color obtaining portion 13 that obtains the color of the pixel included in the image frame, an evaluation score calculating portion 14 that calculates an evaluation score indicating a motion level of the color on the basis of the motion degree for each color obtained by the color obtaining portion, and a moving object detecting portion 15 that detects the moving object on the basis of the evaluation score for each color.

Abstract: An image processing method for reducing noise of an image to be filtered; the method includes a step of creating shrunken image of shrinking ratio r×r (where r is a natural number) by averaging the pixel values for each pixel in a block for every r pixels, a step of obtaining the pixel value in the shrunken image that correspond to the focus pixel in the source image by performing linear interpolation from the shrunken image pixels, a step of determining the noise reduction filter from the focus pixel and reference pixel in the shrunken image, and a step of computing the compensated pixel value of the focus pixel in the source image by weighted sum of each pixel value applying the filter to the focus pixel in the source image and the corresponding shrunken pixel.

Abstract: An image processing device that sequentially generates a synthetic image by joining a first image, which is a single image or constituted by a plurality of the images joined together, and a second image that is input, every time when the second image is input, includes: an overlapping region obtainment portion 14 that obtains a motion vector on the basis of the second image and an image input immediately before the second image among the images constituting the first image, and obtains an overlapping region where the image input immediately before and the second image overlap each other on the basis of the obtained motion vector; a brightness-value conversion factor calculation portion 16 that calculates a brightness conversion factor for converting a brightness value of the first image or a brightness value of the second image on the basis of the brightness value of the first image and the brightness value of the second image in the overlapping region so that a change in the brightness values between the fir

Abstract: Multi-resolution images of a reference image and a target image are generated. Then, whole-range matching is performed on an image of a lower resolution to detect a two-dimensional displacement between the images. Block matching is performed on an image of a higher resolution to detect a displacement at each feature point. The accuracy of motion data is increased by correcting the motion data with an image of a higher resolution by using the previously calculated motion data of the lowest resolution through higher resolutions as an initial value.

Abstract: An image processing device that accumulatively generates a synthetic image by joining a first image, which is a single image or is constituted by a plurality of the images joined together, and a second image that is input, every time when the second image is input, the device comprising: a center-position obtainment portion (11) that obtains position information on a first center point being a center point of each of the images constituting the first image and position information on a second center point being a center point of the second image; and a synthetic image generation portion (12) that obtains the first center point of an image overlapping the second image among the images constituting the first image, and that joins the first image and the second image to generate the synthetic image, using perpendicular bisectors between the obtained first center points and the second center point as a joint between the first image and the second image, based on the obtained position information on the first cent

Abstract: An image data processing method for sharpening a captured image; the method defines a first transformation equation to change the pixel values of a defocused image into a focused image using the coordinates of the defocused image and the focused image as arguments, calculates a DetailedFocus function from the optimal solution of the first transformation equation by giving the pixel values of the focused image and the defocused image as educational data, extracts a predetermined number of important points on the DetailedFocus function, defines a second transformation equation to change the pixel values of the important points of a defocused image into a focused image using the coordinates of the defocused image and the focused image as arguments, calculates a SmartFocus function from the second transformation equation, and produces a focused image from a defocused image using the SmartFocus function.

Abstract: An object detection apparatus that detects an object to be detected captured in a determination image according to a feature amount of the object to be detected preliminarily learned by the use of a learning image, the object detection apparatus including a detector causing strong classifiers to operate in order of lower classification accuracy, continuing processing when the strong classifier has determined that the object to be detected is captured in the determination image, and determining that the object to be detected has not been detected without causing the strong classifier having classification accuracy higher than the aforementioned strong classifier to operate, when the strong classifier has determined that the object to be detected is not captured in the determination image, wherein the strong classifier inputs a classification result of the strong classifier having classification accuracy lower than the aforementioned strong classifier and determines whether the object to be detected is captured

Abstract: A method for capturing an image using an image capturing module with multiple pixels configured in a 2-dimensional matrix and a focal plane electronic shutter function which sequentially slides the exposure timing by constant interval by pixel group which is a unit of predetermined number of the pixels while scanning; the method includes an image data input process for detecting shutter operation, obtaining a reference image while skipping one or more of the pixel groups by sequentially sliding the exposure timing by the constant interval, and obtaining a target image while not skipping pixel groups to be captured, a motion data detection process for calculating the motion data of the target image based on the reference image, and a distortion compensation process for compensating for distortion in the target image according to the motion data.

Abstract: An image processing device includes: processing pixel group determination unit that determines a to-be-processed pixel group from a plurality of frame images of picture signals input based on a first predetermined rule and that determines a processed pixel group from the processed frame image stored based on a second predetermined rule; frequency component processing unit that extracts at least a temporal frequency component from the to-be-processed pixel group and the processed pixel group, that executes predetermined coefficient processing on a temporal high-frequency component of the temporal frequency component and that acquires the executed to-be-processed pixel group; processing completion frame generation unit that generates a processing completion frame image from the executed to-be-processed pixel group; and image output unit that outputs the processing completion frame image and that stores the processing completion frame image in the processed frame storage unit, in which the frequency component pr

Abstract: A motion vector calculation method includes: processing for calculating multi-resolution data up to a predetermined resolution level L from a plurality of inputted image data; and processing for estimating motion vectors per resolution executing a matching step for temporarily calculating motion vectors minimizing an energy function within a predetermined range of the input motion vectors and a smoothing step for finally calculating the motion vectors by averaging the surrounding data. By using the motion vectors of the resolution level L finally calculated as input motion vectors of a further higher resolution level L-1, the processing for estimating motion vectors per resolution is successively repeated on image data of a high resolution, thereby calculating the motion vectors.

Abstract: Multi-resolution images of a reference image and a target image are generated. Then, whole-range matching is performed on an image of a lower resolution to detect a two-dimensional displacement between the images. Block matching is performed on an image of a higher resolution to detect a displacement at each feature point. The accuracy of motion data is increased by correcting the motion data with an image of a higher resolution by using the previously calculated motion data of the lowest resolution through higher resolutions as an initial value.