646 results about "Spatial transformation" patented technology

Methods and systems for automatically generating a mask delineating a region of interest (ROI) within an image containing skin are disclosed. The image may be of an anatomical area containing skin, such as the face, neck, chest, shoulders, arms or hands, among others, or may be of portions of such areas, such as the cheek, forehead, or nose, among others. The mask that is generated is based on the locations of anatomical features or landmarks in the image, such as the eyes, nose, eyebrows and lips, which can vary from subject to subject and image to image. As such, masks can be adapted to individual subjects and to different images of the same subjects, while delineating anatomically standardized ROIs, thereby facilitating standardized, reproducible skin analysis over multiple subjects and/or over multiple images of each subject. Moreover, the masks can be limited to skin regions that include uniformly illuminated portions of skin while excluding skin regions in shadow or hot-spot areas that would otherwise provide erroneous feature analysis results. Methods and systems are also disclosed for automatically registering a skin mask delineating a skin ROI in a first image captured in one imaging modality (e.g., standard white light, UV light, polarized light, multi-spectral absorption or fluorescence imaging, etc.) onto a second image of the ROI captured in the same or another imaging modality. Such registration can be done using linear as well as non-linear spatial transformation techniques.

In a method and apparatus for magnetic resonance imaging of an interconnected region of a human body on the basis of a partially parallel acquisition (PPA) by excitation of nuclear spins and measurement of the radio-frequency signals produced by the excited nuclear spins, a number of spin excitations and measurements of an RF response signal are implemented simultaneously in every component coil of a number of RF reception coils. As a result a number of response signals are acquired that form a reduced dataset of received RF signals for each component coil. Additional calibration data points are acquired for each reduced dataset. A complete image dataset is formed for each component coil on the basis of the reduced dataset for that component coil and at least one further, reduced dataset of a different component coil. A spatial transformation of the image dataset of each component coil is implemented in order to form a complete image of each component coil.

A video coding algorithm supports both lossy and lossless coding of video while maintaining high color fidelity and coding efficiency using an in-loop, reversible color transform. Accordingly, a method is provided to encode video data and decode the generated bitstream. The method includes generating a prediction-error signal by performing intra/inter-frame prediction on a plurality of video frames; generating a color-transformed, prediction-error signal by performing a reversible color-space transform on the prediction-error signal; and forming a bitstream based on the color-transformed prediction-error signal. The method may further include generating a color-space transformed error residual based on a bitstream; generating an error residual by performing a reversible color-space transform on the color-space transformed error residual; and generating a video frame based on the error residual.

A method for displaying in a coherent manner the changes over time of a web site's structure, usage, and content is disclosed. Time tubes are generated by a method of displaying a related series of graphs. Time tubes illustrate changes in a graph that undergoes one or more transformations from one state to another. The transformations are displayed using the length of the cylindrical tube, filling the length of the time tube with planar slices which represent the data at various stages of the transformations. Time tubes may encode several dimensions of the transformations simultaneously by altering the representation of size, color, and layout among the planar slices. Temporal transformations occur when web pages are added or deleted over time. Value-based transformations include node colors, which may be used to encode a specific page's usage parameter. Spatial transformations include the scaling of physical dimension as graphs expand or contract in size. The states of a graph at various times are represented as a series of related graphs. In a preferred embodiment, an inventory of all existing nodes is performed so as to generate a list of all nodes that have existed at any time. This inventory is used to produce a layout template in which each unique node is assigned a unique layout position. To produce each planar slice, the specific nodes which exist in the slice are placed at their respective positions assigned in the layout template. In another aspect, corresponding nodes in planar slices are linked, such as with translucent streamlines, in response to a user selecting a node in a planar slice by placing his cursor over the selected node, or to show clustering of two or more nodes in one planar slice into a single node in an adjacent planar slice.

An image processing apparatus classifies a variation of a target object included in an image from a specific state as one of a plurality of types of attributes, and holds, for each variation attribute, a correction parameter for spatial transformation that corrects the target object to the specific state. The image processing apparatus generates an input image vector by vectorizing at least a partial region of the input image, and determines a variation attribute by detecting a variation of the target object from the specific state in the input image. Then, the image processing apparatus generates a transformed image vector by performing the spatial transformation on the input image vector, using a correction parameter selected based on the determined variation attribute from among the correction parameters held for respective variation attributes.

A method for segmenting a color document into regions of text and halftone discriminates between text and halftone by examining the texture of the document. A color document is digitized, and a color space transform is preferably applied to the digitized document. The texture of the document is identified and a noise reduction step is preferably applied. Bounding boxes (blocks) within the document are identified and then the areas within the bounding boxes are classified as either text or halftone. Two alternative methods are described for examining the document texture. In the first method, a windowing operation is applied to either an intensity image or a color difference image. In the second method, a wavelet transform step combined with Fuzzy K-Mean clustering is applied. Bounding boxes are classified as either text or halftone based upon the relative periodicity of a horizontal or vertical (or both) histogram of each bounding box.

Temporal and spatial scaling of video images including video object planes (VOPs) (117, 118, 119, 405, 415, 420, 430, 520, 522, 524, 526, 532, 542, 705, 730, 750, 760, 780, 790, 805, 815, 820, 830, 850, 860, 880, 890) in an input digital video sequence is provided. Coding efficiency is improved by adaptively compressing scaled field mode video. Upsampled VOPs (450, 490, 522, 542, 750, 790) in the enhancement layer are reordered to provide a greater correlation with the input video sequence based on a linear criteria. The resulting residue is coded using a spatial transformation such as the DCT. A motion compensation scheme is used for coding enhancement layer VOPs (450, 460, 480, 490, 522, 524, 526, 542, 750, 760, 780, 790, 850, 860, 880, 890) by scaling motion vectors which have already been determined for the base layer VOPs (405, 415, 420, 430, 520, 532, 705, 730, 805, 815, 820, 830). A reduced search area whose center is defined by the scaled motion vectors is provided. The motion compensation scheme is suitable for use with scaled frame mode or field mode video. Various processor configurations achieve particular scaleable coding results. Applications of scaleable coding include stereoscopic video, picture-in-picture, preview access channels, and ATM communications.

A method of adaptively inserting a key frame according to video content to allow a user to easily access a desired scene. A video encoder includes a coding mode determination unit receiving a temporal residual frame with respect to an original frame, determining whether the original frame has a scene change by comparing the temporal residual frame with a predetermined reference, determining to encode the temporal residual frame when it is determined that the original frame does not have the scene change, and determining to encode the original frame when it is determined that the original frame has the scene change, and a spatial transformer performing spatial transform on either of the temporal residual frame and the original frame according to the determination of the coding mode determination unit and obtaining a transform coefficient. A keyframe is inserted according to access to a scene based on the content of an image, so that usability of a function allowing access to a random image frame is increased.

A method and apparatus for scalable video and decoding are provided. A method for video coding includes eliminating temporal redundancy in constrained temporal level sequence from a plurality of frames constituting a video sequence input, and generating a bit-stream by quantizing transformation coefficients obtained from the frames whose temporal redundancy has been eliminated. A video encoder for performing the encoding method includes a temporal transformation unit, a spatial transformation unit, a quantization unit, and a bit-stream generation unit. A video decoding method is in principle performed inversely to the video coding sequence, wherein decoding is performed by extracting information on encoded frames by receiving bit-streams input and interpreting them.

Scalable video coding and decoding methods, a scalable video encoder, and a scalable video decoder. The scalable video coding method includes receiving a GOP, performing temporal filtering and spatial transformation thereon, quantizing and generating a bitstream. The scalable video encoder for performing the scalable video coding method includes a weight determination block which determines a weight for scaling. The scalable video decoding method includes dequantizing the coded image information obtained from a received bitstream, performing descaling, inverse spatial transformation, and inverse temporal filtering on the scaled transform coefficients, thereby recovering video frames. The scalable video decoder for performing the scalable video decoding method includes an inverse weighting block. The standard deviation of Peak Signal to Noise Ratios (PSNRs) of frames included in a group of pictures (GOP) is reduced so that video coding performance can be increased.

A system and method for representing a two-dimensional spatial transformation that describes the transformation by an inverse mapped grid data set. The grid data for each coordinate is surface fitted on an array of rectangular patches defined in the output space using numerical techniques. Error analysis determines whether a finer mesh resolution is required for surface fitting. The spatial transformation is then defined by the array of rectangular surface patches and the set of surface coefficients such that the spatial transformation can be executed through evaluation of the surface polynomials. The two-dimensional surface polynomial representation allows the transformation to be easily adjusted for scale changes and zooming and panning effects.

The invention discloses an automatic digging method for remarkable objects of color images. The automatic digging method for the remarkable objects of the color images is characterized by comprising the steps of changing a red green blue (RGB) color space of the images into a hue saturation value (HSV) color space and calculating significance of targets in the aspects of hue, saturation and warm color gain; obtaining a rectangular frame surrounding a significance region according to a significance level image and performing rectangular frame expansion, wherein images surrounded by the expanded rectangular frame serve as follow-up processing images; using an improved watershed algorithm to pre-divide input image contents, using pre-divided super pixel subregions to replace weighted graphs formed by dividing pixel point structural graph, and adopting a maximum flow-minimum dividing strategy to perform division till energy functions are converged to obtain divided images. By means of an automatic image digging technology, the remarkable objects in scene can be quickly and effectively dug, and digging efficiency, the quality and the like are improved remarkably.

A global calibration method of a laser tracking visual guidance measurement system comprises the steps of determining a visual system coordinate system and a laser tracker coordinate system; putting a plane target for at least three times and obtaining plane equations of the target plane, where the plane target is located in the visual system coordinate system and the laser tracker coordinate system at each putting position; and calculating a spatial transformation relation of the visual system coordinate system and the laser tracker coordinate system according to the coefficients of the obtained plane equations. The global calibration method of a laser tracking visual guidance measurement system in the invention realizes the global calibration of the laser tracking visual guidance measurement system by obtaining the plane equations of the target plane in the visual system coordinate system and the laser tracker coordinate system at the at least three positions of the plane target, so as to realize automatic assembly and butt joint of large-sized components, to improve assembly efficiency, and to save assembly cost.

The invention provides a deep learning method for ship detection in high-resolution visible remote sensing images, which comprises the following steps: firstly, reading and preprocessing image data; secondly, extracting the features of the whole image; thirdly, screening out target candidate regions after extracting the abstract features of the image in the convolution layer; fourthly, cutting outthe feature blocks of each target candidate region on the feature map corresponding to the whole image, and using the pooling layer in the region of interest to normalize the sizes of the feature blocks; fifthly, sending the features to the full connection layer to get spatial transformation parameters, and sending the spatial transformation parameters and the features to the spatial transformation layer to get the features after deformation correction; and sixthly, carrying out classification and position correction again on the target candidate regions according to the corrected features. The robustness of the detection method to target rotation and other deformation is enhanced, and the detection effect of ship targets in high-resolution visible remote sensing images is improved. The method can be applied to the detection of ship targets in high-resolution visible remote sensing images, and has broad application prospects and values.

A method of projecting digital information on a real object in a real environment includes the steps of projecting digital information on a real object or part of a real object with a visible light projector, capturing at least one image of the real object with the projected digital information using a camera, providing a depth sensor registered with the camera, the depth sensor capturing depth data of the real object or part of the real object, and calculating a spatial transformation between the visible light projector and the real object based on the at least one image and the depth data. The invention is also concerned with a corresponding system.

Disclosed is a scalable video coding algorithm. A method for video coding includes temporally filtering frames in the same sequence to a decoding sequence thereof to remove temporal redundancy, obtaining and quantizing transformation coefficients from frames whose temporal redundancy is removed, and generating bitstreams. A video encoder comprises a temporal transformation unit, a spatial transformation unit, a quantization unit and a bitstream generation unit to perform the method. A method for video decoding is basically reverse in sequence to the video coding. A video decoder extracts information necessary for video decoding by interpreting the received bitstream and decoding it. Thus, video streams may be generated by allowing a decoder to decode the generated bitstreams, while maintaining the temporal scalability on an encoder-side.

An image synthesis method includes a placement information obtaining step of obtaining placement information about a plurality of images in which adjacent images have a common subject region, and a synthesis step of combining the images using a mapping mode in accordance with the placement information. Prior to performing mapping transformation of the input images, coordinate-space transformation parameters for transforming the coordinate space of one image among the images into the coordinate space of another image are generated. The images are combined based on a given mapping mode and the coordinate-space transformation parameters.

The invention discloses a road warning mark detection and recognition method based on block recognition. The method comprises the following steps of performing HSV (Hue, Saturation and Value) color spatial alternation on vehicle-mounted video data and performing binarization processing; extracting profile information of a binarization image, and judging the shape of the binarization image according to the geometrical characteristics of the binarization image so as to detect a warning mark region; extracting block features and HOG (Histogram of Oriented Gradient) features of the warning mark region; matching out the type of a mark to be recognized by using a mark base as the standard; and obtaining the mark detection result. The method has the advantages that the high mark recognition correct rate can be ensured; and the missing detection is effectively reduced, so that the better mark detection result is obtained. A good application value is realized in the field of intelligent traffic.

The invention discloses a correction method for fisheye image distortion. The method includes the following steps that firstly, a fisheye image of a standardized test grid plate and an actual image of the fisheye image are obtained; secondly, distortion measurement is conducted on the obtained fisheye image, and distortion coefficients are calibrated according to a polar coordinate model; thirdly, correction is conducted according to a correction model, and in order to guarantee correction precision, sectioned correction is conducted according to a central view field, a middle view field and an edge view field during correction; fourthly, pixels obtained after spatial alternation are endowed with corresponding gray values through a bilinear interpolation method so that the gray value of an original position can be restored. Reverse operation is simple, measurement precision is high, and the correction method is easily achieved in hardware systems such as an EPGA and a DSP. The correction method has high practical value.