283 results about "Representative point" patented technology

Provided are a method of efficiently transmitting motion model parameters using temporal correlation between video frames and a video encoding and decoding method and apparatus, in which motion estimation and motion compensation are performed by generating a plurality of reference pictures that are motion-compensated using motion model parameters. Motion model parameters are encoded based on temporal correlation between motion vectors of representative points expressing the motion model parameters, global motion compensation is performed on a previous reference video frame using motion model parameters in order to generate a plurality of transformation reference pictures, and a current video frame is encoded using the plurality of transformation reference pictures.

This invention, by imparting different functions to respective dots of a dot pattern which is displayed on a printed material etc., aims to enable informatization expeditiously by recognizing directionality, on the occasion of informatization from the dot pattern, and to enable checking an error of a layout state of dots, and further, to heighten security. In this connection, in this invention, on a medium surface of a printed material etc., a plurality of lattice dots 4 are disposed in a rectangular shape and set as a block, and they are regularly and continuously disposed, and such a dot that 1 piece of the lattice dots 4 which exists in the block was disposed by being shifted unidirectionally is set as a key dot 2, and by setting said key dot 2 as a representative point, they are disposed at a circumference of the key dot 2, and by setting a center which was surrounded by the lattice dots 4 of 4 points as a hypothetical point, and by setting this as a start point, at an end point which was represented by a vector, a plurality of information dots 3 which have various information recognized are arranged in accordance with a predetermined rule by a dot code generation algorithm to thereby generate a dot pattern 1, and the dot pattern 1 is imported as image data by a camera, and, from a numerical value which was calculated by digitizing this, information, a program are outputted.

A method for clustering documents is provided. Each document is represented by a multidimensional data point. The data points are initially assigned to a respective cluster and serve as their initial representative points. Thereafter, in an iterative process, the data points are clustered among the clusters, by assigning the data points to the clusters based on a comparison measure of each data point with the cluster or its representative point, and a threshold of the comparison measure. Based on this clustering, a new representative point for each of the clusters can be computed. Optionally, overlapping clusters are merged. For the next iteration, the new representative points are used as the representative points. An assignment of the documents to the clusters is output, based on a clustering of the data points in the latest iteration. Multiple batches may be processed, retaining the initial clusters to which the original batch was assigned.

The invention includes a step S1 for inputting into a computer coordinate values on a three-dimensional shape; a step S4 for structuring an environment model that partitions a spatial region, in which a three-dimensional shape exists, into a plurality of voxels of rectangular solids, and stores each position; and a step S5 for setting and recording a representative point and an error distribution thereof, within the voxel corresponding to the coordinate value. If there is no data in a previous measurement position, position matching is performed in a fine position matching step S7 so as to minimize an evaluation value regarding the distances between adjacent error distributions by rotating and translating a new measurement data and error distribution for the environment model for a previous measuring position, or rotating and translating an environment model for a new measuring position, relative to an environment model for a previous measuring position.

The invention provides a fine structure deformation monitoring method based on ground three-dimensional laser scanning. The method comprises steps that, a GPS or a total station is utilized to arrange artificial target points, and a target point monitoring net is established; point cloud data at a surface of a structure at different moments are acquired by utilizing a ground three-dimensional laser scanner; data acquired by the target point monitoring net and data acquired by the ground three-dimensional laser scanner are converted to a same reference coordinate system; the data in the same reference coordinate system are utilized to determine and fit a geometrical model of the structure; point cloud data segmentation is carried out according to the geometrical model of the structure obtained through fitting, the point cloud data segmentation is divided into multiple parts, and each part is fitted into one representative point; change of coordinates and vectors of the representative points obtained through point cloud data fitting at different moments are contrasted to carry out fine structure deformation analysis. The fine structure deformation monitoring method can realize monitoring on fine structure deformation.

An object region data describing method of describing information about the object region in an image over a plurality of frames determines the region of a target object in an image using an approximate function that approximates the trajectory obtained by arranging, in the direction of frame advance, one representative point of an approximate figure for the object region and the difference values for determining the other representative points and describes information about the object region using a parameter for the function.

A map information correction device, a map information correction method, a program, an information providing device using the program, and information acquisition device, in which displacement of an object caused by an error between different maps is correctable. An information acquisition device (600) has a positional information receiving section (601) for receiving information on an object including outline position reference information for referring a position near the object, the information on an object corresponding to a digital map database (A507) on the information provider side; an outline position specifying section (602) for specifying an outline position; a representative position extraction section (604) for extracting a representative point from the outline position; and an inner shape correction section (605) for performing interpolation processing by the use of information on positions in digital map database (A507) and in digital map database (B603) and correcting at least either the position or the shape of the object so that it fits a second map.

A 3D-object detecting apparatus may include a detection-image creating device configured to detect a 3D object on an image-capture surface from an image captured by an image-capture device and to create a detection image in which a silhouette of only the 3D object is left; a density-map creating device configured to determine the 3D objects spatial densities at corresponding coordinate points in a coordinate plane on the basis of the detection image and mask images obtained for the corresponding coordinate points on the basis of virtual cuboids arranged for the corresponding coordinate points and to create a density map having pixels for the corresponding coordinate points such that the pixels have pixel values corresponding to the determined spatial densities; and a 3D-object position detecting device that detects the position of the 3D object as a representative point in a high-density region in the density map.

The invention discloses a fuzzy clustering based image retrieval method. The fuzzy clustering based image retrieval method comprises the following steps of S11, establishing a characteristic value library for images in an image library and numbering the images ; S12, selecting N images with image separation distances larger than a distance threshold value A1 from the image library and conducting first-time classification on the residual images to form N categories of image sets; S13, conducting the step S12 on the image sets with image quantities larger than a quantity threshold value in the N categories of image sets till the image quantity of each image set is smaller than the quantity threshold value and obtaining M representative points; S14, partitioning all images in the image library into the image sets represented by highest-similarity-level representative points according to similarity levels between the images and the M representative points; S15, conducting characteristic valuing on input images to be retrieved, respectively calculating the similarity levels between the input images and all representative points and selecting a plurality of highest-similarity-level representative points to perform retrieval. The fuzzy clustering based image retrieval method narrows a retrieval range on the basis that retrieval efficiency is ensured, and reduces retrieval working amount.

In an image taking apparatus according to the present invention, focus adjustment is performed based on image information in a focus adjustment area within a captured image in response to a movement instruction from a user as well as a focus adjustment mode switching instruction from the user. When detecting an instruction to switch to a position non-fixed focus adjustment mode, an initial position of the focus adjustment area in the position non-fixed focus adjustment mode immediately after mode switching is determined based on a position of a representative point of the focus adjustment area immediately before the mode switching.

The invention discloses a large-data clustering algorithm based on a cloud computing platform. Primitive data is pre-processed; data are divided into M sub-data and distributed to M Map functions and local clustering is carried out on the sub-data; clusters with the same key are combined; if the number R of practical clusters is smaller than the number k of clusters, the number c of representative points and a constriction factor a are regulated and clustering is carried out again until termination conditions are achieved. if a new data set is generated, local clustering is carried out according to judgment conditions that the number K of new data source centers is larger than the obtained number K of clusters before updating or the number of new data source points is larger than the number of data source points before updating. According to the large-data clustering algorithm based on the cloud computing platform, the parallel computing capacity of a high-performance clustering system of cloud computing is used for solving the problem that mass data need to be processed in clustering, and therefore a relation of the data can be rapidly and efficiently dug up.

A theoretical expression of a workpiece (W) the curved surface of which is measured by a measuring probe (110) equipped with a stylus (111) is specified, a measuring area where the measurement is executed on the measuring surface of the workpiece (W) is determined, and the axis angles of the stylus (111) are determined based on the coordinate values and the normal vector of a representative point determined in the measuring area.

The invention provides a method for rapidly and accurately focusing and scanning pathological section tissue on the basis of an image collection device. The method comprises steps as follows: (1), the image collection device is prepared; (2), a representative point on the pathological section tissue is selected, a focus point of the representative point is focused, a focal length of each representative point is calculated, and a virtual plane is fit; (3), a first scanning line of X, Y and Z axes is taken as a movement track, and the pathological section tissue is divided into at least two lines for scanning; (4), a camera starts capturing images, the movement track of Z axis is continuously corrected according to the clarity trend of a front line and a rear line obtained by the camera in the process, that is, the focal length is changed; and (5), scanning is completed. The invention further provides a scanner based on the method. By means of the method, a large quantity of focus points is reduced, a rough virtual plane is obtained, and the movement track of the Z axis is continuously corrected according to image data in the scanning process, so that the problems of low focusing speed, inaccuracy and the like are well solved.

The invention relates to a real-time predicting method for output of a wind electric field, which aims at providing a real-time predicting method for output of a wind electric field, which has high predicting precision, strong practicality and short predicting reaction time. The output of the wind electric field is predicted through predicting the wind speed of the wind electric field, and the invention aims to provide reliable basis for a power network to make a schedule plan. The technical problem for solving the problem is as follows: the real-time predicting method for output of a wind electric field is characterized by comprising the following steps of: a. simulating a large-scale circulation field and a meteorological field by a number weather reporting mode according to the real-time refreshed meteorological grid point initial data, and deducing the wind resource state of the representative point; b. sending the wind resource state obtained from the step a to an output calculating model; c. deducing the output of the whole wind electric field by using the output calculating model, and furthermore, deducing the output of a wind electric field group. The invention is mainly used in the wind energy utilizing field, predicts the electric generation amount of a large wind electric generating field, and provides basis for the power network schedule.

A lane marker recognizing apparatus which recognizes stud-type lane markers from acquired road image includes a candidate region extracting means for extracting a region having the possibility of being an image portion of the lane marker, from the road image, as a lane marker candidate region, a real space representative point calculating means for determining a representative point of the lane marker candidate region according to a predetermined condition, and calculating a real space position corresponding to the representative point as a real space representative point, a grouping means for forming one group of the real space candidate points having a relative distance within a predetermined range set according to standards on the lane marker, and a lane marker position recognizing means for recognizing the position of the lane marker based on the real space representative point formed into one group by the grouping means.

A method of simplifying the arithmetic operation in a global motion compensation process approximates the motion vector field of the whole image without using many parameters. Motion vectors in the global motion compensation are found by the interpolation and/or extrapolation of the motion vectors of a plurality of representative points 602, 603 and 604 having particular features in the spatial distance thereof. Since the shift operation can be substituted for the division for synthesizing a predicted image of global motion compensation, the processing using a computer or a dedicated hardware is simplified.

The invention discloses a coding method based on the spatial characteristics of vector data primitives, the steps of which are: determining the coded quadtree grid reference; Row number and column number, perform quadtree row and column encoding; select the point that best represents the primitive, and generate a representative point code according to the relative position of this point in the quadtree grid reference; base on the regional characteristics of the primitive Contour encoding; and then encode the orientation of the primitive data based on the direction of the head point of the primitive relative to the representative point; finally, according to the unified coding rules, combine the four sub-codes into the spatial feature code of the primitive, as the unique identification of the primitive . The invention comprehensively considers multiple spatial features of vector data elements, solves the influence of area re-division on coding, eliminates the interference of artificial classification, and improves the efficiency and accuracy of coding vector data primitives.