63 results about "Transformation geometry" patented technology

A projection system uses a transformation matrix to transform a projection image p in such a manner so as to compensate for surface irregularities on a projection surface. The transformation matrix makes use of properties of light transport relating a projector to a camera. If the resolution a camera is lower than that of a projector within said projection system, then the transformation matrix will have holes where image data corresponding to a projector pixel will have been lost. In this, case, new image are generated to fill-in the holes.

The invention discloses a fine-grained emotion polarity prediction method based on a hybrid attention network, and aims to overcome the problems of lack of flexibility, insufficient precision, difficulty in obtaining global structure information, low training speed, single attention information and the like in the prior art. The method comprises the following steps: 1, determining a text context sequence and a specific aspect target word sequence according to a comment text sentence; 2, mapping the sequence into two multi-dimensional continuous word vector matrixes through log word embedding;3, performing multiple different linear transformations on the two matrixes to obtain corresponding transformation matrixes; 4, calculating a text context self-attention matrix and a specific aspect target word vector attention matrix by using the transformation matrix, and splicing the two matrixes to obtain a double-attention matrix; 5, splicing the double attention matrixes subjected to different times of linear change, and then performing linear change again to obtain a final attention representation matrix; and 6, through an average pooling operation, inputting the emotion polarity into asoftmax classifier through full connection layer thickness to obtain an emotion polarity prediction result.

A font style transfer and transformation method based on Manchu character picture belongs to the field of image processing, which solves the problems of greatly reducing the time and labor cost spentin processing a large number of Manchu character images. The main points are that the manchu font style transfer model is trained step by step: the manchu character training data set is used for training the model to realize the manchu character learning; a mandarin font style migration data set is use to train that model again to realize the mandarin font style migration learning, and the effectis to save a large amount of time and labor cost.

A power transmission and transformation project three-dimensional-aided preliminary design system comprises a data storage unit, a power transmission and transformation project typical model base unit, a data exchange interface, a three-dimensional rendering engine unit and a spatial analysis unit. The data storage unit is used for storing and managing spatial data and non-spatial data needed to be referenced in designing power transmission and transformation projects of various types, the power transmission and transformation project typical model base unit is used for storing power transmission and transformation equipment three-dimensional models of various types and with representative significance by combining actual power transmission and transformation project design, the data exchange interface is used for inputting and outputting power transmission line design tower optimized arranging software data formats, the three-dimensional rendering engine unit is used for creating, loading and rendering three-dimensional scenes, and the spatial analysis unit is used for providing spatial measuring and spatial analysis functions. The power transmission and transformation project three-dimensional-aided preliminary design system created by applying the three-dimensional GIS (geographic information system) technology integrates data of various types needed at a preliminary design stage, comprehensively judges various factors influencing power transmission line route selection and transformer station address selection by stacking of the data of various types, and overcomes defects of poor up-to-date state, poor visuality and the like of topographic maps in conventional methods.

The invention provides a monocular visual positioning method based on inverse perspective projection transformation. The technical scheme employed in the invention is that: an attitude sensor and a camera are fixed together and arranged on a wheeled vehicle; and images shoot in driving are subjected to the following processes: step 1, carrying out inverse perspective projection transformation on an image sequence; step 2, obtaining transformation matrixes between adjacent images through calculating; and step 3, determining a driving locus curve of the wheeled vehicle. The method allows high accurate positioning results to be obtained by assisting to position the wheeled vehicle through real time attitude information obtained by the attitude sensor, and the perspective effect to be eliminated through carrying out inverse perspective projection transformation on the images, so the precision of wheeled vehicle positioning can be further improved.

The invention discloses a method for rapidly identifying an environmental risk of a power transmission and transformation project. The method comprises the following steps of: modeling spatial data information which comprises a map, a line route, a sensitive area position boundary and the like according to an object-oriented data model; establishing corresponding spatial database and attribute database; querying related data of a sensitive area meeting the condition in the attribute database by adopting a standard Structure Query Language (SQL); screening a spatial entity identifier meeting the condition; entering the spatial database; and searching a corresponding spatial entity according to the identifier so as to realize automatic identification statistics, analysis and early warning, and planning and comparison of the sensitive area, and further complete the environmental risk assessment and environmental optimization of power transmission and transformation project construction.

The invention relates to an AADL2TASM model transformation method, which is used for transforming an AADL model into a TASM model. KM3 is used for describing the meta-model of the source language AADL and the meta-model of the target language TASM, and aimed at a thread component, a dispatcher, mode conversion and a scheduler in the AADL model, a transformation method for the thread component, a transformation method for the dispatcher component, a transformation method for the mode conversion and a transformation method for the scheduler component are respectively used for transforming the AADL model into the TASM model.

The invention discloses a quick point cloud registration method based on curved surface fitting coefficient features. Curvature mean differences of neighborhoods of different sizes are compared, points whose differences exceed a set threshold are selected to serve as key points, and key point candidate points are adaptively selected according to the differences. Multiple neighborhoods are selected at the key point for curved surface fitting, and a curved surface coefficient serves as a feature descriptor for the key point. Through comparing distances between key point feature descriptors, a key point pair with the smallest distance is selected to serve as an initial corresponding relation. A transformation matrix obtained through the initial corresponding relation is used for adjusting positions and orientations of the corresponding relation for basic coincidence, a distance threshold is set, and corresponding relations whose distances are larger than the threshold are removed. Then, a clustering method is used for enabling the corresponding relations to be uniformly distributed, a covariance matrix for the corresponding relations after optimization is calculated, and singular value decomposition is then carried out on the covariance matrix to obtain a final transformation matrix. The method of the invention has the advantages of quick registration, high precision and good anti-noise ability.

The map display device of the present invention includes: a model-view transformation matrix generating unit (102c) for generating a model-view transformation matrix of 4 rows and 4 columns for converting the global coordinates of a 3-dimensional object into a viewpoint coordinate system; a model-view transformation matrix changing unit (102d), when performing coordinate transformation by multiplying the model-view transformation matrix from the left side of the coordinates, change the value of the first row, the second column, and the third row, the second column of the model-view transformation matrix to 0, and On the one hand, when performing coordinate conversion by multiplying the model-view transformation matrix from the right side of the coordinates, change the value of the 2nd row, 1st column and the 2nd row, 3rd column of the model-view transformation matrix to 0; the model view The conversion unit (102e) converts the global coordinates into the viewpoint coordinate system using the changed model-view conversion matrix.

The invention discloses a tire inner wall texture identification method based on a locating block. The tire inner wall texture identification method based on the locating block includes steps: obtaining a coordinate of the locating block of an image frame, and transforming an original image at any angle into a standard front view by employing a transformation model; obtaining a static texture description by employing a static texture model through image enhancement; and performing texture identification on sampled images. According to the method, the standard front views can be obtained by transformation of the images with different angles by employing a transformation method, the fact that static texture presents the same characteristic in time and space is researched, a dynamic problem is changed into a static problem, a group of static texture models are obtained for the static texture, matching and identification with a database are performed so that rapid texture matching is realized, and the effect is obvious by employing voting and marking with the combination of various methods.

The invention provides a constraint heteroscedasticity linear discriminant analysis method for language identification, which relates to a method for the dimension reduction and decorrelation of high-dimension feature vectors. The method is characterized in that MFCC features are extracted from voice signals; the MFCC features of continuous M frames are selected and placed in parallel so as to obtain a cepstrum matrix; the cepstrum matrix is expanded according to rows to form super vectors; the mean and covariance of the super vectors are calculated by the block; a transformation matrix is calculated by the block by an iteration method; the super vectors are transformed by the block by use of the transformation matrix; and each block is subjected to dimension reduction and decorrelation treatment so as to obtain new feature vectors. The method has the advantage of obtaining the feature vectors of which the correlation among dimensions is removed, along with small amount of calculation, high discriminant property and low dimension, and can be used for language identification.

The present invention includes a rotation based transformation method for preserving data privacy. A rotation based transformation method according to an embodiment of the present invention divides a given data set into a plurality of data subsets, applies a plurality of different rotation matrixes to the plurality of divided data subsets, respectively, to rotate the plurality of data subsets, and re-concatenates the plurality of rotated data subsets to generate a rotated data set. The rotated data set is released and used for data clustering. The rotation based transformation method is effectively used when the impact of the AK-ICA attack is mitigated. Meanwhile, when the two parties are related to the present invention, the two parties need to agree on parameters, such as the number of data subsets. It is preferable that the data subsets be divided into equal sizes. Further, rotation-unification work that removes relative rotation deviation between the plurality of different rotation matrices can be performed in order to provide valid clustering between the data subsets rotated using the different rotation matrixes.

The present invention belongs to the wireless communication method technical field and relates to a spectrum sensing method based on the alpha-divergence of information geometry. The method includes the following steps that: a cognitive radio model is established, a noise environment is simulated, so that N noise covariance matrices Rk are generated, and one of the N noise covariance matrices Rk is adopted as a unit to be detected R; the Riemann mean overlineR of the N noise covariance matrices Rk is calculated through using a natural gradient descent algorithm, a geometric distance L betweenthe detection unit R and the Riemann mean overlineR is calculated; n geometric distances L are sequenced in a descending order, and a threshold value T is calculated; simulation is performed, so thata plurality of unknown signal covariance matrices are generated, and the alpha-divergence of the information geometry is adopted to solve distances D between the signal covariance matrices and a knownsignal matrix in manifold; and the threshold value T is compared with the distances D, if D is larger than T, signals to be sensed exist, otherwise, only noises exist. The method of the invention hasthe advantages of simple and practical calculation process, high spectrum sensing efficiency and high spectrum sensing stability.

The invention provides a Karhunen-Loeve Transform (KLT) matrix training method for a video intra-frame coding belonging to the video coding field in a signal treatment. The invention firstly picks-up a residual signal frequency spectrum under each intra-frame prediction model, then groups a residual sample and gets a candidate KLT matrix by using a conventional method, optimizes each candidate KLT matrix by using a low frequency subband preferential replacement policy and calculates the optimized transformation performance, and selects an optimization KLT matrix with best transformation performance as a transformation matrix of the residual signal under the prediction model. The invention avoids great calculation amount for real-timely training the KLT matrix during the coding, reduces an additional bit overhead for showing the training matrix, and the trained matrix has very good transformation performance.

The invention provides a fracturing transformation method for a blocking lake basin shale oil reservoir layer. The method comprises the following steps of S1, obtaining key data according to existingwell logging data explanations, and determining the source reservoir configuration relations according to the longitudinal overlapping relation of different kinds of rock in well logging lithologic sections; S2, according to existing rock mechanics tests, ground stress tests and natural fracture distribution data, formulating a fracturing technology optimum selecting principle for reservoirs withdifferent source reservoir configuration relations; S3, dividing preponderant lithofacies according to the lithologic sections and the cloud-ground ratio, and formulating an optimum selecting principle for the classification of the preponderant lithofacies, a fracturing liquid anti-expanding system and proppants through the combination of existing core anti-swelling experiments and rock board embedding experiment data. By means of the fracturing transformation method for the blocking lake basin shale oil reservoir layer, according to the differences in the rock types and the stratum source reservoir configuration relations, the individualized fracturing liquid system can be optimally selected, the fracturing pump casting procedures can be optimized, and the fracturing transformation effectis improved.

The invention discloses an intelligent extracting method for engineering characteristic indexes in paragraph contents of a word document of a transmission and transformation project. The method sequentially comprises the following steps of creating an engineering characteristic index library; extracting the outline of the engineering characteristic indexes; and extracting the paragraph contents of the engineering characteristic indexes. The created index library serves as a foundation database for extracting the engineering characteristic indexes; and when the characteristic indexes of the transmission and transformation project are extracted, the outline of the characteristic indexes is extracted at first, then the paragraph contents are extracted, and after the paragraph contents are extracted, the required characteristic indexes of the transmission and transformation project are obtained. By using the extracting method, the required characteristic index information of the transmission and transformation project can be automatically extracted from a designed file, and the work efficiency of an expert on a design review is improved.

The invention relates to an encryption method of the shifting rule and the transposition rule of a double circulation queue. The method comprises the following steps: a plaintext information matrix is taken; a random integer sequence is created; a matrix transformation frequency is input by user; a random number is taken out from the random integer sequence and is rounded; circular queue shifting and transposition of one row or one column of the transformation matrix are carried out according to a set remainder rule; a cipher text matrix is obtained after repeated transformations; finally, all transformation operation processes are recorded as sub keys which are arranged in turn to obtain a key. The encryption method, which belongs to a classic permutation encryption, adopts basic shifting operation and realizes quick enciphering process and deciphering process by means of less resource; meanwhile, the method solves the problems of the prior shifting encryption method such as low encryption intensity, low speed and poor operability; moreover, the encryption method which has the advantages of high encryption speed, simple implementation, high encryption intensity, excellent safety and a wide application scope is suitable to be used in software implementation and hardware implementation.

The present invention provides a security assessment method for an AADL model based on deterministic stochastic Petri nets. The method comprises: building an AADL architecture model of a hierarchicalstructure; then, developing a new model transformation method to transform the AADL security model into a DSPN model, and adding new semantics, including a relationship between an operational state ina composite component and a failure state, a relationship between the operational state and the hazard, and a relationship between a recovery event and a repair event, and error detection, to the AADL security model, wherein the new model transformation method is a transformation method for simplifying error propagation between components and the transformation method for simplifying dangerous behaviors, so that the security model is transformed into a DSPN model with a smaller number of states to calculate the probability of the hazard; and finally, in combination with the severity level ofthe hazard, calculating the risk of the hazard, and performing security assessment.

The invention relates to a manual web page optimizing method during web page transformation. The method comprises the steps as follows: (1) optimizing range is selected, and a web page of the optimizing range is pre-browsed; (2) each continuum of browsed web page requiring manual optimizing transformation is selected, and corresponding transformation rule is set towards each continuum; and (3) when the automatic transformation of the web page is carried out on a transformation engine, the corresponding optimizing transformation rule is preferentially applied to the area which is provided with the optimizing transformation rule. The optimizing unit is generally a specific Scope URL, a website and a specific route under the website and also can be a specific web page, etc. The optimizing transformation rule generally comprises folding, hiding, newline, inserting and / or font, etc. The invention overcomes the disadvantage that the transformation rule which is set uniformly in advance is not fit for corresponding website or web page, and the invention improves the transformation effect and ensures that the transformed web page is more practicable and nicer.

The invention discloses a transformation matrix-based transformer three-phase model building method. The method comprises the following steps of establishing an original admittance matrix Yp of a three-phase transformer, and according to primary and secondary side connection modes of the transformer, modifying a transformation ratio of the matrix Yp; according to the primary side connection mode,determining a primary side voltage transformation matrix Cp; according to a primary and secondary side wiring mode, determining Csp and Cps; according to mark numbers of groups, converting a basic secondary side voltage transformation matrix Css through a transformation matrix to obtain a secondary side voltage transformation matrix Cs and modifying Cps; according to sub-matrixes of the voltage transformation matrixes, forming a voltage transformation matrix C and a current transformation matrix G of the transformer; and according to the matrix Yp and the transformation matrix, deriving a nodeadmittance matrix YTn of the three-phase transformer. The method solves the problem that a relation matrix method is not suitable to obtain node admittance matrixes of transformers of certain connection groups, further reduces the software writing workload and the error probability, and improves the accuracy of obtaining a three-phase model of the three-phase transformer.

The invention provides a geometric body cross section demonstration device comprising a transparent geometric body container and a needle tube injector. One side of the geometric body container is provided with an air hole, and the opposite side is provided with a liquid inlet hole. The liquid inlet hole is internally provided with a protruding short conduit. The short conduit is flexibly connected with the injection port of the needle tube injector. Liquid is arranged in the needle tube injector. The liquid volume in the geometric body container is directly controlled by pushing and pulling the piston of the needle tube injector, and the liquid surface is enabled to present the shape of the cross section by changing the position and the angle of the geometric body container so that control is facilitated, the teaching efficiency can be enhanced, the liquid is absorbed out of the geometric body container into the needle tube injector after completion of teaching, and thus the device is easy to carry and is not liable to damage.

The invention provides a folding ruler which is formed by serially connecting a plurality of ruler bodies through rivets and mainly used for teaching. When the folding ruler is used, the length of the folding ruler can be adjusted according to using needs, and then a teacher can utilize the folding ruler to perform measurement calculation and plotting on the blackboard. Besides, by changing folding modes of the folding ruler, the teacher can make a variety of geometric shapes to perform explanation for students. Therefore, the folding ruler has the advantages of performing calculation and geometric shape transformation, and facilitating improvement on understanding of the students.

The invention discloses a man-machine interaction interface graph rendering method based on a transformation matrix. By means of a the matrix composed of translation, rotation and scaling parameters attached to a texture, the texture can conduct translation, rotation, scaling and other operation through the transformation matrix, accordingly various kinds of complex graph rendering of a man-machine interaction interface can be achieved, and complex graph operation, like translation, rotation and scaling, achieved by a traditional man-machine interface at high cost can be achieved at low cost. The invention provides a man-machine interaction interface graph rendering system based on the transformation matrix, a texture rasterization unit is added, rotation, translation and scaling of the texture are supported, a user can self-define the transformation matrix, functions of the traditional man-machine interface are finished at extremely low cost, and accordingly speed and effects of the man-machine interface are increased and improved.

The invention discloses a method for predicting investment for power transmission and transformation projects for power grids on the basis of particle swarm algorithms. The method includes analyzing key indexes for predicting the investment for construction costs for the power transmission and transformation projects; analyzing unit projects of the power transmission and transformation projects; analyzing systems for predicting the investment for the power transmission and transformation projects. The method has the advantages that investors can accurately master conditions of the construction costs of the new construction projects in initial construction periods, review staffs can reasonably and quickly review conditions of construction costs for specific initial design schemes, accordingly, influence of subjective factors in construction cost management procedures can be reduced, and the purpose of providing bases for investment decision making can be achieved; the key indexes which can affect prediction of the investment for the power transmission and transformation projects start to be determined from decision making phases of the power transmission and transformation projects, historical project data are mined, the key unit projects of the power transmission and transformation projects are determined, the systems for predicting the investment for the power transmission and transformation projects are constructed by the aid of the particle swarm algorithms, and accordingly theoretical bases can provided for controlling the construction costs for the power transmission and transformation projects and reasonably planning the investment for the projects by all levels of competent departments and related units for the power grids.

The invention discloses a multi-antenna cooperative spectrum sensing method based on information geometry and a differential evolution clustering algorithm. The multi-antenna cooperative spectrum sensing method comprises the following steps: dividing a received signal Xj into two clusters according to a clustering mode, i.e., respectively carrying out the principal component analysis of the sum, and correspondingly obtaining and constructing covariance matrixes R1 and R2 corresponding to the sum and the construction; extracting signal features according to information geometry; in an noise environment, performing clustering processing and principal component analysis processing on the P noise signal matrixes to obtain a sum, calculating a corresponding Riemann mean value, and mapping the covariance matrixes R1 and R2 obtained in the step S2 to a statistical manifold; and taking the Riemannian mean value sum as a reference point, and analyzing the distances d1 and d2 between the pointsto obtain a signal feature T. According to the technical scheme, a threshold does not need to be calculated, and whether the main user exists or not is judged through the differential evolution clustering algorithm, and higher judgment precision is achieved.

The invention relates to an inverse transformation method and system of perspective images. In license plate recognition application at the entrance of a parking lot, license plate images photographed by a photographing device are perspective images which are deflected and deformed due to limitations of actual environment. According to the inverse transformation method and system of the invention, an X-coordinate and Y-coordinate inverse transformation formula can be obtained through derivation according to a lens imaging geometric law of an object, and a geometric law when an included angle exists between the normal of a photographed surface of the object and an imaging target surface, and as a result, the coordinate point of each pixel after transformation can be obtained, and therefore, the perspective images can be reduced to similar images of the original object, and the success rate of vehicle license plate recognition can be improved.

The invention relates to a result post-processing oriented OpenFOAM finite volume analysis result data format and transformation method thereof. The step includes: finding an OpenFOAM grid file and aresult data file; matching expressions containing data information in the grid file and the result data file by using regular expressions; establishing and completing mapping relationships between alldata files; defining a data storage form (syntax, semantics, and a format of an XML file); and generating and exporting data files with the mapping relationships according to the XML file format. Theinvention provides a syntax of XML format, realizes automatic transformation of OpenFOAM result data file to the XML format, and provides an interface file for post-processing of OpenFOAM simulationcalculation results and secondary development. The file structure is clear and intuitive, and facilitates reading and data extraction and transmission.

When deducing a homography matrix representing a geometrical relationship between two images, an image processing apparatus obtains a transformation matrix for four feature points in each of the two images based on reference coordinates. More specifically, the apparatus obtains the area of a triangle formed by three out of the four feature points, and calculates projective parameters of transformation matrices from the area ratios. The apparatus calculates the homography matrix between the images by the product of one of the two obtained matrices and an inverse matrix of the other.