Abstract: An embodiment of a method for computing pyramids of input images (I) in a transformed domain, e.g., for search and retrieval purposes, includes:—arranging input images in blocks to produce input image blocks,—subjecting the input image blocks to block processing including: transform into a transformed domain, subjecting the image blocks transformed into a transformed domain to filtering, subjecting the image blocks transformed into a transformed domain and filtered to inverse transform implementing an inverse transform with respect to the previous transform into a transformed domain, thus producing a set of processed blocks. The set of processed blocks, which is recomposeable to an image pyramid, may be used, e.g., in detecting extrema points in images in the pyramid, extracting a patch of given size around the extrema points detected, and processing the patch to obtain local descriptors such as SIFT descriptors of a feature.

Abstract: An embodiment is a method for detecting image features, the method including extracting a stripe from a digital image, the stripe including of a plurality of blocks; processing the plurality of blocks for localizing one or more keypoints; and detecting one or more image features based on the one or more localized keypoints.

Abstract: An embodiment relates to a method for the detection of texture of a digital image, including providing a raw data image of the image by means of Bayer image sensors, determining noise in at least a region of the raw data image and determining the texture based on the determined noise without using a high pass or low pass filter.

Abstract: An embodiment relates to a method and an image processor for reducing chroma noise in digital-image data. An embodiment performs noise reduction in the color-filter-array domain prior to demosaicing in order to prevent spreading of noise in subsequent stages of the image-processing pipeline. Peaks in the CFA data are attenuated in order to prevent any undesired color cast. Specifically, any correction to a certain pixel is made in accordance with the amplitude of digital gains applied, as well as with the local luminance and the contribution of the current color channel to the local luminance. In this manner, corrections are restricted to image areas that are subject to high digital amplification, that are comparatively dark, and that are not dominated by the current color channel.

Abstract: A method and system for filtering an image frame of a video sequence from spurious motion, comprising the steps of dividing the image frame and a preceding image frame of the video sequence into blocks of pixels; determining motion vectors for the blocks of the image frame; determining inter-frame transformation parameters for the image frame based on the determined motion vectors; and generating a filtered image frame based on the determined inter-frame transformation parameters; wherein the image frame is dived into overlapping blocks.

Abstract: An embodiment relates to a method for the detection of texture of a digital image, including providing a raw data image of the image by means of Bayer image sensors, determining noise in at least a region of the raw data image and determining the texture based on the determined noise without using a high pass or low pass filter.

Abstract: Image defects in digital images are easily detectable by the human eye but may be difficult to detect in a computer-implemented fashion. In an embodiment of a digital-image-acquisition device, defects are removed on the CFA domain before color interpolation takes place. In order to allow cancellation of couplets of defective pixels, a two pass embodiment is presented. Such embodiment presents methods and systems that can remove both couplets and singlets without damaging the image. The system includes a ring corrector that detects a defect in the ring of pixels that surround a central pixel, a singlet corrector that detects and corrects the central pixel and removes a couplet if the ring corrector is activated, whereas if the ring corrector is switched off, the singlet corrector only removes singlets, and a peak-and-valley detector that avoids overcorrection by avoiding correcting signal peaks or valleys in case of spikes or drops in signal.

Abstract: In an embodiment, focusing an image-capture device such as, e.g., a camera including an optical system displaceable in opposite directions (A, B) via a focusing actuator, is controlled by evaluating a scale factor for the images acquired by the device. An accumulated value of the variations of the scale factor over a time interval (e.g., over a number of frames) is produced and the absolute value thereof is compared against a threshold. If the threshold is reached, which may be indicative of a zoom movement resulting in image de-focusing, a refocusing action is activated by displacing the optical system via the focusing actuator in the one or the other of the opposite focusing directions (A or B) as a function of whether the accumulated value exhibits an increase or a decrease (i.e., whether the accumulated value is positive or negative).

Abstract: A panning device for processing relative motion vectors and absolute motion vectors obtained from a video sequence, includes: a panning filter module, such as a high-pass IIR filter, for subjecting relative motion vectors to panning processing, an adder module for adding the relative motion vectors subjected to panning in the panning filter module to absolute motion vectors to obtain respective summed values of motion vectors, a clipping module for subjecting the summed values of motion vectors obtained in the adder module to clipping according to a selected cropping window for obtaining final output absolute motion vectors, a first leak integrator arranged after the panning filter module, and a second leak integrator arranged after the clipping module.

Abstract: A method of processing digital images by transforming a set of pixels from a three-dimensional space to a normalized two-dimensional space, determining a membership class and membership class level of each pixel in the set of pixels, and selectively modifying colors of pixels in the set of pixels based on the determined membership classes and membership class levels.

Abstract: According to an embodiment, a sequence of video frames as produced in a video-capture apparatus such as a video camera is stabilized against hand shaking or vibration by:—subjecting a pair of frames in the sequence to feature extraction and matching to produce a set of matched features;—subjecting the set of matched features to an outlier removal step; and—generating stabilized frames via motion-model estimation based on features resulting from outlier removal. Motion-model estimation is performed based on matched features having passed a zone-of-interest test confirmative that the matched features passing the test are distributed over a plurality of zones across the frames.

Abstract: An embodiment of a method for computing pyramids of input images (I) in a transformed domain, e.g., for search and retrieval purposes, includes:—arranging input images in blocks to produce input image blocks,—subjecting the input image blocks to block processing including: transform into a transformed domain, subjecting the image blocks transformed into a transformed domain to filtering, subjecting the image blocks transformed into a transformed domain and filtered to inverse transform implementing an inverse transform with respect to the previous transform into a transformed domain, thus producing a set of processed blocks. The set of processed blocks, which is recomposeable to an image pyramid, may be used, e.g., in detecting extrema points in images in the pyramid, extracting a patch of given size around the extrema points detected, and processing the patch to obtain local descriptors such as SIFT descriptors of a feature.

Abstract: An embodiment relates to a method for color processing of an input image, the method including the steps of low-pass filtering of the input image to obtain a low-pass component, high-pass filtering of the input image to obtain a high-pass component, processing the input image for edge detection to obtain edginess parameters, and performing a color-space transformation of the input image based on the low-pass component, the high-pass component, and the edginess parameters.

Abstract: An embodiment includes a method and an apparatus for the generation of a visual story board in real time in an image-capturing device including a photo sensor and a buffer, wherein the method includes the consecutively performed steps: starting the recording of a video, receiving information on an image frame of the video, comparing the information on the received image frame with information on at least one of a plurality of image frames wherein the information on the plurality of image frames has previously been stored in the buffer, storing the information on the received image frame in the buffer depending on the result of the comparison, and finishing the recording of the video.

Abstract: A method and system for filtering an image frame of a video sequence from spurious motion, comprising the steps of dividing the image frame and a preceding image frame of the video sequence into blocks of pixels; determining motion vectors for the blocks of the image frame; determining inter-frame transformation parameters for the image frame based on the determined motion vectors; and generating a filtered image frame based on the determined inter-frame transformation parameters; wherein the image frame is dived into overlapping blocks.

Abstract: A method of processing and device configure to process digital images to enhance image quality and correct motion blur. A number N of images of a scene are captured with an exposure time T. An order of sharpness of the images is determined and the sharpest image is used as a reference image for generating an output image.

Abstract: Apparatus and methods to unwarp at least portions of distorted, electronically-captured images are described. Keypoints, instead of an entire image, may be unwarped and used in various machine-vision algorithms, such as object recognition, image matching, and 3D reconstruction algorithms. When using unwarped keypoints, the machine-vision algorithms may perform reliably irrespective of distortions that may be introduced by one or more image capture systems.

Abstract: A method of processing digital images by transforming a set of pixels from a three-dimensional space to a normalized two-dimensional space, determining a membership class and membership class level of each pixel in the set of pixels, and selectively modifying colors of pixels in the set of pixels based on the determined membership classes and membership class levels.

Abstract: An embodiment relates to a method and an image processor for reducing chroma noise in digital-image data. An embodiment performs noise reduction in the color-filter-array domain prior to demosaicing in order to prevent spreading of noise in subsequent stages of the image-processing pipeline. Peaks in the CFA data are attenuated in order to prevent any undesired color cast. Specifically, any correction to a certain pixel is made in accordance with the amplitude of digital gains applied, as well as with the local luminance and the contribution of the current color channel to the local luminance. In this manner, corrections are restricted to image areas that are subject to high digital amplification, that are comparatively dark, and that are not dominated by the current color channel.

Abstract: An embodiment relates to a method for the detection of texture of a digital image, including providing a raw data image of the image by means of Bayer image sensors, determining noise in at least a region of the raw data image and determining the texture based on the determined noise without using a high pass or low pass filter.