222 results about "Feature modeling" patented technology

A method for simulating a physical system using finite element techniques, wherein two or more distinct models corresponding to distinct regions within the modeled system are solved, each with a corresponding evaluator. Nodes which lie on the boundaries between the models may have different values corresponding to the different models. When a particular model is solved, the evaluator for that model is used to obtain the appropriate values for each of these common nodes. In one embodiment, a first model is defined, then a region corresponding to a particular feature within the system is carved out of it. A finite element model corresponding to the feature is then inserted into the region. The finite elements may be adapted to share nodes on the boundaries between them.

The invention provides a modeling method of the characteristics of population space-time dynamic moving based on multisource data fusion. The modeling method comprises: A. inputting map data, mobile phone locating data and floating vehicle data into a system and managing data organization according to requirements; B. establishing a spatial analysis model of the characteristics of population moving based on the mobile phone locating data and the floating vehicle data; C. applying the spatial analysis model of the characteristics of the population moving to carry out multisource data fusion of the map data, the mobile phone locating data and the floating vehicle data to obtain integrated information of the characteristics of the population moving; and D. analyzing the characteristics of various population moving according to the integrated information of the characteristics of the population moving and publishing an analyzed result by the geographic information system. The modeling method can acquire data of urban population space-time dynamic distribution and moving characteristics with large data amount, high quality and space-time characteristics, obtain basis of accurate population distribution and population moving characteristics, and provide decision-making supports for urban planning, land use planning, transportation planning and the like.

The invention discloses a phishing website detection method based on uniform resource locator (URL) classification. Firstly, a modeling is performed for URL characteristics, a method specific to a domain name imitation phenomenon of the characteristics is provided for calculating the similarity between a suspicious domain name and a protected domain name by a dynamic programming thought, in order to collect phishing URL high frequency suspicious character features, a suspicious character extraction algorithm based on a generalized suffix tree is provided, then on the basis of characteristic modeling, a support vector machine (SVM) algorithm is utilized to perform classified training for experimental training set, a SVM classification model is obtained after the training, the SVM classification model is used for classifying the URL to be detected, and a server for detecting a phishing website updates the current SVM classification model according to a specific online incremental learning strategy.

The invention provides a data parallel processing method. The data parallel processing method comprises the following steps that 1, a main management node receives data and acquires the incidence relation of the data; 2, the main management node calculates allocatable GPUs and GPU work loads of work computing nodes; 3, the main management node partitions the data and distributes the partitioned data to all the work computing nodes; 4, the work computing nodes perform parallel processing on the received data and transmit processing results back to the main management node; 5, the main management node merges the results and then outputs the results. The data parallel processing method has the following advantages that a master-slave architectural pattern is adopted to be used for high-performance large-scale data parallel processing, operation stage partition is performed on specific operations converted by application programs according to DNA feature modeling, node granularity grade operation deployment is performed according to a partition result, and the execution efficiency of a parallel task of data flow in a single node is improved by adopting a thread parallel optimization mechanism and fully utilizing multiple computing kernels.

The invention discloses a method for identifying benign and malignant lung nodules based on multi-dimensional information. The method comprises the following steps: 1, representing a three-dimensional nodule in a two-dimensional way; 2, building a feature model; 3, building a fuzzy classifier; 4, evaluating the classifying performance. The method has the beneficial effects that accurate feature modeling is crucial to the identification of benign and malignant lung nodules. More objective bases are laid for the identification of the benign and malignant lung nodules by adopting imaging diagnosis features, common shapes and textual features in image processing, and patient information, feature extraction is performed on a two-dimensional image generated on the basis of a helical scanning technology, and a novel method is adopted for feature modelling, so that the extracted features are more accurate. In the method, a fuzzy C-means (FCM) clustering algorithm is adopted for identifying benign and malignant status of a suspected nodule, and a probability indicating the suspected nodule is benign or malignant is given, so that the method is more accordant with the thinking mode of a doctor.

The invention provides a sparse sample-oriented focus type Web information extraction system and method. The sparse sample-oriented focus type Web information extraction system includes: a webpage interaction module for providing extraction template definition and structuralized extraction result search service; an extraction engine module for providing functions of similar webpage acquisition, sample feature modeling, feature selection, and information extraction; and a data service module for providing a relationship type data service and a non-relationship type data service for the front end and the back end of the system. Based on a small number of samples, high-efficient information extraction can be performed, and the structuralized information can be extracted out form the fields to which different samples belong.

The invention discloses a Web service recommendation method based on user preference feature modeling. The Web service recommendation method based on user preference feature modeling comprises the steps that (1) Web services are collected from a Web service portal, the service roles, targets, processes and marks of the collected Web services are labeled and registered to a service repository in a Web service platform, and a Web service description document is built; (2) the historical service call information of each user is collected from the Web service platform, a historical call information document about the service roles, targets, processes and marks of the users is generated, and a user preference document is built; (3) by calculating the multi-dimensional preference similarity among the users and carrying out weighted fusion, the former N preference similarity neighbors of the users are generated, and N>=1; (4) the Web services most frequently called by the former N preference similarity neighbor users are sequenced, and a final service recommendation list is generated. The Web service recommendation method is suitable for the Web services described by various languages, such as services described by WSDL, services described by OWL-S, and the web services described through the method like the text language, and has universality.

The invention discloses an intelligent identification method and device for a baby sleep position. A video analysis and mode identification method is adopted to identify the baby sleep position so as to timely discover important baby events like a baby kicks out a quilt, the face of the baby is covered by clothes, or the baby sleeps on the stomach. The method consists of three parts, namely sample feature modeling, real-time feature analysis and alarm judgment. In the sample feature modeling, textural features and SIFT (Scale Invariant Feature Transform) features of a sample image are analyzed, a sample feature template base is generated by means of feature fusion, furthermore, the features of a set monitoring area are analyzed while the real-time features are analyzed, the sleep position is identified in combination with the sample feature template base, the alarm type is judged, and then the alarm information is output. Due to the intelligent identification method and device, a guardian does not have to monitor the baby all the time by videos or observation on site, especially when the guardian sleeps soundly at night, the important baby events can be effectively and intelligently detected, identified and warned early and timely.

The invention discloses a real-time-robust pedestrian detection method aiming at a specific scene, comprising the following steps: using a camera to carry out video collection in the specific scene; preprocessing the collected video information at real time by adopting a background difference method to obtain the video information of movable objects in the processed video; taking each movable object in the preprocessed video information as center to demarcate a sliding window; and detecting the image information in the sliding window by a support vector machine pedestrian classifier to match a pedestrian. The feature modeling training of the support vector machine pedestrian classifier comprises the following steps: 1) feature extraction, in which a head and shoulder training sample library is formed by the histogram features at head and shoulder gradient direction and local binaryzation model features; and 2) sample training, in which the head and shoulder training sample library formed by the histogram features at head and shoulder gradient direction and local binaryzation model features is placed into the support vector machine for training so as to obtain the support vector machine pedestrian classifier. The pedestrian detection method has the characteristics of high accuracy and instantaneity.

The invention is suitable for the field of intelligent process design, and provides an MBD model based processing feature identification and modeling method. The method comprises: based on PMI extraction, obtaining product manufacturing information; during identification of a geometric attribute of a processing feature, synthesizing one or more pieces of geometric element information into the processing feature according to a topological relation of geometric elements, wherein the geometric attribute of the processing feature includes geometric attributes of the geometric elements; during identification of a process attribute of the processing feature, obtaining the process attribute of the feature from the topological relation of the geometric elements and labeled information associated with the geometric elements according to the geometric attribute of the processing feature; and finishing processing feature modeling according to the geometric attribute of the processing feature and the process attribute of the processing feature. According to an embodiment of the invention, a part model is analyzed from a process perspective, the processing feature is defined, and PMI information is converted into the process attribute of the processing feature. The method increases the utilization rate of a three-dimensional CAD model and has important significance for exciting the positivity of three-dimensional CAD application.

The invention discloses a music similarity detection method based on mixed characteristics and a Gaussian mixed model. According to the basic thought, the method comprises the steps of using a gamma-tone cepstrum coefficient for conducting similarity detection, and using weighting similarities of various characteristics as a final detection result; providing a modulation spectrum characteristic based on a frame shaft, using the characteristic for representing a music long-time characteristic, and using the combination of the long-time characteristic and a short-time characteristic as the input of modeling in the next step; using the Gaussian mixed model for conducting modeling on the music characteristics, firstly, utilizing a dynamic K mean value method for conducting initialization on the model, then, using an expectation-maximization algorithm for conducting model training, obtaining accurate model parameters, and finally using a log-likelihood ratio algorithm for obtaining the similarities between the pieces of music. According to the music similarity detection method, the obtaining of the music characteristics is more sufficient and thorough, the accuracy degree of music recommendation is improved, the characteristic vector dimensionality can be reduced, the information memory content of a music database is reduced, and the accuracy degree of the music recommendation is improved.

The invention relates to a method for controlling a flexible satellite based on a feature model, which is characterized in that the time dimension, the sampling time and the parameters M and m are determined according to the kinetic equation of the flexible satellite; the coefficient range of the feature model is determined according to each obtained variable; the parameters of the feature model are identified by utilizing a gradient method; and a control law is designed according to the coefficients of the feature model obtained through the identification, and the attitude angle of the flexible satellite is controlled through the kinetic equation that the control law is fed back to the flexible satellite. The method has the advantages that the time dimension and the sampling period of the flexible satellite are introduced; the change rate of the flexible satellite is depicted; the bottleneck problem of the feature modeling of the flexible satellite is solved; the expression of the parameter range of the feature model of the flexible satellite is provided; the parameter property of the feature model is qualitatively researched; the boundary of the parameters of the feature model is relative to the sampling period, the modeling error, the system order and the change rate of the system from the given parameter range; and the theoretical foundation of the self-adaptive control based on the feature model is laid for the flexible satellite. The method is suitable for the feature model of the attitude kinetics of an aircraft so as to lay the foundation for the aircraft based on the attitude control of the feature model.

The invention discloses a real-time performance test method which is applied to real-time performance tests of an embedded system. The real-time performance test method includes following steps: performing time characteristics modeling on application program source codes in an upper computer; using a time characteristics model obtained in the modeling process to complete instrumentation operation for the source codes; downloading the source codes onto a lower computer after the source codes are compiled and interlinked, and starting an application program through the upper computer; collecting test data during the running period of the application program, and returning data stored in a certain data container in real time to the upper computer according to preset transmission rules; obtaining a real-time performance test report of the application program after data analysis. The real-time performance test method achieves the real-time performance tests for the embedded real-time application program, and has the advantages of being fully automatic, small in error and detailed in result during the test process. The invention further discloses a system which uses the real-time performance test method.

A particle filter target tracking improvement method based on a vision attention mechanism includes steps: 1, initializing input; and 2, modeling a feature and filtering particles based on a visual saliency mechanism, selecting an optimum feature according to saliency order of features, modeling the feature and filtering mismatching particles to obtain an optimum particle set finally. The features comprise a color feature, a texture feature and a movement feature. A target position is calculated according to the optimum particle set, tracking results are output, and simultaneously the saliency order of the features is updated according to the optimum feature. The particle filter target tracking improvement method is a multi-feature target tracking method for simulating a human vision mechanism, combines the color, texture and movement features, and can guarantee real-time performance, accuracy and robustness.

The invention provides a convolution neural network-based music recommending system and method. The system comprises a music user modeling module for collecting historical behavior data of a music user and constructing a preference model of the music user; a music feature modeling module for obtaining a regression model; and a recommendation algorithm module for finding music objects matched withthe preference of the music user through the regression model and recommending the music objects to the music user. According to the system provided by the invention, deep learning is applied to the recommending system, semantic differences between song features and audio signals are effectively compensated and the problems such as "cold start" and the like in collaborative filtering are avoided at the same time, so that the accuracy of the recommending system is increased; and the contradiction between low training efficiency and a high timeliness requirement is solved by adopting a convolution neural network and the historical behavior information of the user and audio acoustic features are added to the model, so that the recommendation results are more in line with the preference requirements of the user and the user experience of the recommending system is improved.

The invention relates to a navigational positioning method based on sky polarization distribution model matching. The method comprises the following steps: taking a positive direction of a mobile robot as a 0-degree reference direction, photographing the sky above the mobile robot in real time by employing a polarization camera, acquiring a real-time sky polarization distribution false color image through calculation and synthesis; performing local feature modeling on the sky polarization distribution false color image; performing stable feature point extraction and feature matching on the local features of the sky polarization distribution false color image, and obtaining an affine transformation relationship between the sky polarization distribution model diagrams; and calculating the position and navigational direction of the mobile robot. According to the method, the polarization distribution of the whole sky is not required to be obtained; the local feature information of the sky polarization distribution is utilized, so that the navigational direction of the mobile robot can be acquired from the sky polarization distribution, and the position of the mobile robot can be acquired; and the method is not required to depend on prior knowledge and is high in environmental adaptivity and high in accuracy.

The invention provides a modeling method for dynamic machining features of a complex curved surface and belongs to the technical field of CAD/CAM. The modeling method for the dynamic machining features of the complex curved surface comprises the steps that firstly, according to a selected machining scheme, machining intermediate state features corresponding to all machining stages in the curved surface machining scheme are defined; secondly, curved surface machining area division is conducted on each intermediate stage feature according to selected machining devices, cutting parameters and machining optimization objects of different machining stages, so that curved surface machining sub-features are obtained. According to the modeling method for the dynamic machining features of the complex curved surface, FunctionBlock is used as an information carrier of the curved surface dynamic machining features, when manufacturing resources are changed, an event triggering-responding mechanism of the FunctionBlock is utilized, a defined curved surface feature modeling method in the FunctionBlock is automatically called, and the intermediate state features of curved surface machining and curved surface machining sub-features contained by the intermediate state features are updated in real time. By the adoption of the modeling method for dynamic machining features of the complex curved surface, the curved surface machining features can be dynamically updated according to the change of the manufacturing resources, and the modeling method is the key technology for achieving complex curved surface self-adaption optimized machining.

The invention relates to a Shanghai dialect phonetic recognition information processing method, which includes steps that: 1) a voice input device inputs Shanghai dialect signals; 2) a preprocessing module preprocesses the input Shanghai dialect phonetic signals; 3) a feature extraction module extracts feature parameters reflecting signal features; 4) a training module performs preprocessing and feature parameter extraction on training phonetic signals input by users for a plurality of times to obtain feature vector parameters and then a feature modeling module builds a reference model base for training voice; 5) a recognition module carries out similarity comparison on feature vector parameters of the input voice and models in the reference module base and outputs the input model with the highest similarity as a recognition candidate result; 6) a postprocessing module performs phonetic knowledge processing on the recognition candidate result in step 5) to obtain the final recognition result; and 7) the final recognition result is output through a voice output device. Compared with the prior art, the Shanghai dialect phonetic recognition information processing method has the advantages of being high in recognition speed and the like.

The invention provides a test based static analysis misinformation eliminating method capable of enhancing usability of a software static analysis technology and reducing time and labor cost in artificially confirming and testing a static analysis report. A testing technology is adopted to acquire procedural information which is mutually verified with target paths and target defects related in the static analysis report to confirm the defects or eliminate misinformation. The test based static analysis misinformation elimination method includes reading static analysis report files stored in a metadata interchange format based on an extensible markup language, and analyzing information with the target defects; subjecting each specified defect to static analysis warning, performing continuous concrete execution, symbolic execution and constraint solving on a programs by adopting an idea of mixed execution, and modeling and acquiring the runtime information according to features with different defects during the course; adopting the runtime information acquired in the previous step to confirm whether the defects occur or not, or judging that the static analysis warning is the misinformation; iterating till all the static analysis warnings are processed.

The invention relates to a surveillance video exceptional event detection method based on deep learning and dynamic clustering. In a characteristic extraction stage, a deep learning network PCA (Principal Component Analysis) Net is applied, a video is trained to learn a corresponding network filter, low-layer pixel optical flow characteristics are converted into high-layer semantic motion characteristics through a deep network, and meanwhile, motion areas in a video are screened to remove a spatial-temporal sampling block which only contains background information. In a characteristic modelingstage, a nonparametric model based on two-layer clustering is applied to carry out modeling of characteristic vector space, a vector opposite-direction combination method is adopted in a vector combination stage, finally, a K-means clustering algorithm is applied for clustering vectors in a dictionary set into one series of event clusters, and an exceptional event is judged according to Euclideandistance between a test vector and an event cluster central vector. By use of the method, characteristic vector offset caused by addition can be effectively avoided, and an exceptional event detection rate is improved.

The invention discloses a method for making a feature modeling system on the basis of extensive markup language (XML) in a unigraphics (UG) environment. The method includes following steps that features of a part are classified according to processing and manufacturing angles of the part; relations and connecting and locating relations among different features are summarized according to whether corresponding relations among all features exist or not so that expression of a feature information model is simplified; detailed feature information is described according to feature classifications and feature relations of the part, and a unique feature information model is established; the feature information model is described in an XML format, and a document type definition (DTD) of an XML document is established for cooperating a subsequent computer-aided process planning (CAPP) system to read XML document data; as for the feature modeling system on the basis of UG, a geometric model of the part is converted to an information model in VisualC++6.0 environment by a secondary development tool of a UG system, and a *.txt format document corresponding to the part is finally output. By means of the method, manufacturing information is clearly expressed by hierarchical description, and a developer is allowed to define own elements as required to describe the feature information model.

The invention discloses a video emotion recognition method based on time sequence multi-model fusion modeling, and the method comprises the steps: selecting a data set in a video emotion database as atraining data set, and carrying out the preprocessing of the training data set; constructing a convolutional neural network model based on a feature sampling structure according to the preprocessed training data set; constructing a long-short-term memory network model based on an attention mechanism according to the video spatial feature sequence extracted by the convolutional neural network model; and fusing the convolutional neural network model and the long-term and short-term memory network model to obtain a video emotion recognition model. According to the video emotion recognition method based on time sequence multi-model fusion modeling provided by the embodiment of the invention, the accuracy of video emotion recognition can be effectively improved through the video emotion recognition model constructed by fusing the time sequence feature modeling and other models.

The invention discloses a depth-information-aided particle filter tracking method. A camera like Kinect camera with a depth sensor is used for collecting images to obtain RGB (red-green-blue) and depth data. Particle filter tracking combining depth information is implemented by selecting a tracking target in a first frame to obtain tracking region window size, performing feature modeling on the tracking target according to RGB information of the selected target, a center point depth value and whole-region depth data, and continuing modeling of a next frame of images in the same way; then, calculating the window size of a description region of each particle according to different depth values of each particle at different positions and a depth value of the first frame so as to obtain region features; modeling in the same way as the first frame, comparing features of the next frame of images with those of the first frame so as to obtain optimal particle regions finally. The depth-information-aided particle filter tracking method has the advantage that target tracking accuracy and robustness are improved according to collected RGB-D (RGB-depth) information, namely, color and depth information.

The invention discloses a short-time and long-time feature modeling fusion-based environmental sound recognition method and device. According to the invention, a model cascaded fusion method is adopted, so that short-time and long-time information can be utilized in a whole identification process. According to the technical schemes of the invention, the method includes two stages. According to the first stage, pre-classification is performed on sliding windows based on short-time features and by using the modeling of the Gaussian mixture model (GMM); confidence judgment is performed on the classification results of the GMM; a result with high confidence is directly adopted as a final classification result; and when lower confidence appears, re-classification is carried out based on long-time features. According to the second stage, based on analysis on a GMM classification result confusion matrix, classes easy to be confused are found out; and a support vector machine (SVM) classification model between the classes is trained; and re-classification is carried out by using a support vector machine (SVM). The probability score of the Gaussian mixture model used in the modeling process of the second stage is added to the long-time features, so that the probability score and the long-time features are together adopted as the input of the SVM.