Abstract: A method includes receiving a video signal that comprises a time series of images of a face of a human, wherein the images in the time series of images comprise a set of landmark points in the face, applying tracking processing to the video signal to reveal variations over time of at least one image parameter at the set of landmark points in the human face, generating a set of variation signals indicative of variations revealed at respective landmark points in the set of landmark points, applying processing to the set of variation signals, the processing comprising artificial neural network processing to produce a reconstructed PhotoPletysmoGraphy (PPG) signal, and estimating a heart rate variability of a variable heart rate of the human as a function of the reconstructed PPG signal.

Abstract: A method includes receiving a video signal that comprises a time series of images of a face of a human, wherein the images in the time series of images comprise a set of landmark points in the face, applying tracking processing to the video signal to reveal variations over time of at least one image parameter at the set of landmark points in the human face, generating a set of variation signals indicative of variations revealed at respective landmark points in the set of landmark points, applying processing to the set of variation signals, the processing comprising artificial neural network processing to produce a reconstructed PhotoPletysmoGraphy (PPG) signal, and estimating a heart rate variability of a variable heart rate of the human as a function of the reconstructed PPG signal.

Abstract: A method includes receiving a video signal that comprises a time series of images of a face of a human, wherein the images in the time series of images comprise a set of landmark points in the face, applying tracking processing to the video signal to reveal variations over time of at least one image parameter at the set of landmark points in the human face, generating a set of variation signals indicative of variations revealed at respective landmark points in the set of landmark points, applying processing to the set of variation signals, the processing comprising artificial neural network processing to produce a reconstructed PhotoPletysmoGraphy (PPG) signal, and estimating a heart rate variability of a variable heart rate of the human as a function of the reconstructed PPG signal.

Abstract: This invention concerns a system and a method for the management of delimited areas distributed on a territory. A predisposition of a plurality devices distributed on the territory is provided, each one capable of monitoring a predetermined area, each distributed device being of an “embedded” type and includes: a local “embedded” processor connected to at least one video camera positioned nearby and from which it acquires the images that it processes according to a “deep learning” module and “Computer Vision” technique. The deep learning model is based on a CNN (computer neural network) for the detection of vehicles occupying spaces and the identification of areas that can be occupied, both in delimited areas with a defined layout and in areas with an undefined layout, the data processed being sent to a central unit via an Internet network in one example.

Abstract: A method includes receiving a video signal that comprises a time series of images of a face of a human, wherein the images in the time series of images comprise a set of landmark points in the face, applying tracking processing to the video signal to reveal variations over time of at least one image parameter at the set of landmark points in the human face, generating a set of variation signals indicative of variations revealed at respective landmark points in the set of landmark points, applying processing to the set of variation signals, the processing comprising artificial neural network processing to produce a reconstructed PhotoPletysmoGraphy (PPG) signal, and estimating a heart rate variability of a variable heart rate of the human as a function of the reconstructed PPG signal.

Abstract: An embodiment of a method of recognizing finger detection data in a detection data map produced by a touch screen includes converting the data from the x, y, z space into a three-descriptor space including: a first coordinate representative of the number of intensity peaks in the map, a second coordinate representative of the number of nodes (i.e., pixels) absorbed under one or more of the intensity peaks. A third coordinate may be selected as the angular coefficient or slope of a piecewise-linear approximating function passing through points having the numbers of nodes absorbed under the intensity peaks ordered in decreasing order over said intensity peaks, which permits singling out finger data with respect to non-finger data over the whole of the touch screen.

Abstract: In an embodiment, a first individual image and a second individual image constituting an encoded stereoscopic image, for example in JPEG format with respective levels of encoding quality and united in a multiple-image file, for example of the Multiple-Picture Object (MPO) type. The second level of encoding quality is lower than the first level of encoding quality. During decoding, the first individual image encoded with a first level of encoding quality and the second individual image encoded with a second level of encoding quality lower than the first level of encoding quality are extracted from the multiple-image file, then using information of the first extracted individual image for enhancing the second extracted individual image.

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 of a method of recognizing finger detection data in a detection data map produced by a touch screen includes converting the data from the x, y, z space into a three-descriptor space including: a first coordinate representative of the number of intensity peaks in the map, a second coordinate representative of the number of nodes (i.e., pixels) absorbed under one or more of the intensity peaks. A third coordinate may be selected as the angular coefficient or slope of a piecewise-linear approximating function passing through points having the numbers of nodes absorbed under the intensity peaks ordered in decreasing order over said intensity peaks, which permits singling out finger data with respect to non-finger data over the whole of the touch screen.

Abstract: In an embodiment, a first individual image and a second individual image constituting an encoded stereoscopic image, for example in JPEG format with respective levels of encoding quality and united in a multiple-image file, for example of the Multiple-Picture Object (MPO) type. The second level of encoding quality is lower than the first level of encoding quality. During decoding, the first individual image encoded with a first level of encoding quality and the second individual image encoded with a second level of encoding quality lower than the first level of encoding quality are extracted from the multiple-image file, then using information of the first extracted individual image for enhancing the second extracted individual image.

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: According to the novel method, roto-translational and zooming parameters describing spurious motion effects are determined by exploiting any of the many block matching algorithms commonly used for motion estimation for calculating a motion vector for all or for a selected number of blocks of pixels of the current frame that is processed. Some of the so calculated motion vectors are not taken into account for estimating spurious motion effects. The roto-translational and zooming parameters describing what is considered to be spurious global motion between a current frame and the precedent frame of the sequence, are calculated by processing the selected motion vectors of blocks of pixels of the frame through a recursive procedure that includes computing error values and readjusting the roto-translational and zooming parameters based on the error values.

Abstract: A method for associating with a digital image a class of a plurality of predefined classes having respective models, the method including the phases of dividing the digital image pixel by pixel into one or more regions belonging to a set of predefined regions that differ from each other on account of their type of content, the division being effected by establishing whether or not a pixel of the image belongs to a respective region on the basis of an operation of analyzing the parameters this pixel, the analysis operation being carried out by verifying that the parameters satisfy predefined conditions and/or logico-mathematical relationships of belonging to the respective region, acquiring from the digital image divided into regions information regarding the regions that are present in it, comparing this information with at least one model characterizing a respective class of said plurality, and associating with the digital image a class on the basis of the comparison phase.

Abstract: A digital image processing method includes extracting chromatic information of an image taken by an image taking device and related to a human subject; detecting visually interesting regions; and exposure correcting of the taken image by normalizing a grey scale of the taken image based on the visually interesting regions. Advantageously, the method includes recognizing areas corresponding to the skin of the subject, these areas being used as the visually interesting regions for the exposure correction step.

Abstract: The system carries out conversion of digital video signals organized in blocks of pixels from a first format to a second format. The second format is a format compressed via vector quantization. The vector quantization is performed by means of repeated application of a scalar quantizer to the pixels of said blocks with a quantization step (Q) determined in an adaptive way according to the characteristics of sharpness and/or brightness of the pixels.

Abstract: The present invention sets out to make available a method for reducing noise in an image sequence. This method can be implemented in an acquisition device such as a digital video camera or the like. The aim of this invention is attained with a method for filtering a sequence of digital images in CFA format.

Abstract: According to the novel method, roto-translational and zooming parameters describing spurious motion effects are determined by exploiting any of the many block matching algorithms commonly used for motion estimation for calculating a motion vector for all or for a selected number of blocks of pixels of the current frame that is processed. Some of the so calculated motion vectors are not taken into account for estimating spurious motion effects. The roto-translational and zooming parameters describing what is considered to be spurious global motion between a current frame and the precedent frame of the sequence, are calculated by processing the selected motion vectors of blocks of pixels of the frame through a recursive procedure that includes computing error values and readjusting the roto-translational and zooming parameters based on the error values.

Abstract: A method for transforming a matrix representation of pixels of an image into a vector representation of the image involves partition of the image into contiguous homogeneous regions by applying a watershed type morphologic algorithm to the matrix representation of pixels and identifying the contours of the contiguous homogeneous regions by applying a vector code contouring algorithm of the chain-code type to every region. By selecting quantization level of the gradient, improved image quality, dimensions and noise are achieved.

Abstract: A method of processing digital source images, each represented by pixel matrices, to obtain from two or more source images, representing one and the same real scene and acquired by utilizing different exposure levels, a final digital image capable of reproducing the real scene with an exposure latitude greater than that of each of the source images. The method, which can be advantageously used in digital still cameras, produces the final image by combining the source images with the help of a weighted average constructed pixel by pixel. Thanks to a special filtering to which the weighting coefficients are subjected before the weighted mean operation, the method obtains a final image in which the source images are harmoniously combined with each other.

Abstract: A method of obtaining a depth map from a single digital image of a three-dimensional scene containing geometric objects is provided that includes the identification in the image of at least one horizontal plane or one vertical plane of the three-dimensional scene and the assignment to every point of the image forming part of the identified plane, or to each one of the identified planes, of a depth level according to its distance from, respectively, a predetermined horizontal or vertical reference straight line of the image. In particular, the identification of at least one plane includes the identification in the image of at least two vanishing lines corresponding to two parallel lines in the plane and the assignment of a depth level made in accordance with the angular coefficients of the identified vanishing lines. Also included is a completely automatic method for obtaining a depth map from a single digital image of any three-dimensional scene.