408 results about "Rough set" patented technology

The invention discloses an improved visual attention model-based method of a natural scene object detection, which mainly solves the problems of low detection accuracy rate and high false detection rate in the conventional visual attention model-based object detection. The method comprises the following steps of: (1) inputting an image to be detected, and extracting feature saliency images of brightness, color and direction by using a visual attention model of Itti; (2) extracting a feature saliency image of a spectrum of an original image; (3) performing data sampling and marking on the feature saliency images of the brightness, the color, the direction and the spectrum and an attention image of an experimenter to form a final rough set information table; (4) constructing attribute significance according to the rough set information table, and obtaining the optimal weight value of the feature images by clustering ; and (5) weighing feature sub-images to obtain a saliency image of the original image, wherein a saliency area corresponding to the saliency image is a target position area. The method can more effectively detect a visual attention area in a natural scene and position objects in the visual attention area.

The invention relates to a power transformer fault diagnosis method which combines the extension theory and the rough set theory, belonging to the electric power main equipment fault diagnosis technical field. The invention completes primary reduction classification of the attribute condition needed by various fault types according to a rough set attribute reduction method and then establishes a matter element model of transformer fault diagnosis; DGA testing data is taken as a transformer fault diagnosis attribute set; a transformer standard fault mode is taken as a transformer fault diagnosis decision-making set; various fault degrees are calculated by means of an extension correlation function; moreover, fault inclusion-exclusion rule is defined to determine a transformer fault. The power transformer fault diagnosis method carries out analysis through taking a certain transformer as an example with the diagnosis result according with the practical situation; seventy six pieces of transformer DGA information are collected and fault diagnosis is carried out by means of the method, thereby obtaining higher diagnosis correct rate as compared with IEC three-ratio method.

The invention discloses an anomaly detection method for various kinds of intrusion. The method comprises the following steps of: 1) pre-processing an original data set, identifying a complete request message, and dividing network connection through service type to extract relevant characteristics; 2) by analyzing the characteristics of all kinds of attack by a characteristic extraction unit, and by using application layer information during consideration of relevant fields on the head of a data packet, extracting three characteristics, namely basic characteristics, flow characteristics and content characteristics; 3) by using an attribute reduction algorithm based on a discernibility matrix, processing attributes of a great number of extracted data characteristics, deleting redundant attributes in the attributes to obtain a reduced attribute set, extracting data from original training data according to the reduced attribute set to obtain new training data, and transmitting the new training data to a support vector machine (SVM) module for training and classification; and 4) by using a multi-classification SVM method based on a binary tree, classifying minimum attribute sub-sets after reduction of a rough set to realize a quick classification function of intrusion detection.

The invention discloses a Hadoop-based fast neighborhood rough set attribute reduction method. The method comprises the following steps: a, establishing a distributed platform based on the Hadoop; b, defining a neighborhood rough set; c, generating a candidate set; d, calculating the importance of each attribute; e, selecting the attribute with the largest importance and adding the attribute into the candidate set; f, judging whether a stop condition is met or not; g, storing conditions selected by characteristics. The method is based on the Hadoop distributed platform to analyze the parallelization of a parallel data mining algorithm so as to realize the parallelization of a neighborhood rough set attribute reduction algorithm; the time complexity of the parallelized attribute reduction is greatly lowered, the output of an intermediate result in the performing intermediate process is greatly reduced, and the analysis efficiency of large-scale data is improved, so that numerous and varied mass data are converted into available data with information and business values, thereby completing mining and analysis optimizing of data.

The invention discloses a rough set-based radar radiation source signal identification method, relates to the technical field of signal identification and solves the problem of large calculated amount because the least square needs to be calculated and an optimal initial clustering center needs to be determined when the radar radiation source signal is identified by the conventional rough K-mean value method. The method comprises the following steps of: firstly, acquiring a pulse description word of a radar radiation source signal sample; secondly, determining a clustering number and the initial clustering center of the rough K-mean value by using rough set theory; thirdly, acquiring the centre of RBF neural network hidden layer neurons by using the rough K-mean value so as to acquire an RBF neural network structure; and finally, inputting the sample description word of the radar radiation source signal to be identified into the RBF neural network, and acquiring the identification result to finish the identification of the radar radiation source signal. The method of the invention is suitable for the identification of the radar radiation source signal.

The invention provides a method and a system for monitoring vibration of a wind driven generator. In the method, firstly, based on a large amount of historical data, a vibration characteristic value range rule base is built through a data mining method on the basis of a rough set; secondly, a vibration characteristic value range is predicted by the real-time operating data of the wind driven generator according to the extracted rules in the rule base, and the threshold of the vibration characteristic value is calculated; and finally, the real-time data of the operation characteristic of the wind driven generator and the threshold of the vibration characteristic value are compared so that the judgment of failure warning is made. In the method and the system, different operating data combinations are considered, so that the failure warning rate of monitoring the vibration of the wind driven generator can be greatly reduced; and at the same time, the vibration characteristic value range rule base is represented in a displayable and explainable multiplex rule mode, is easy to understand, and is convenient to automatically or manually maintain.

The invention provides an intelligent fault diagnosis method based on a rough Bayesian network classifier, which comprises the following steps: using standard fault feature data as a fault diagnosis condition attribute set, using a standard fault mode as a fault diagnosis decision attribute set, and adopting a rough set principle to construct an original fault diagnosis information table T1; adopting the minimum entropy method to carry out discrete processing on various continuous fault diagnosis condition attribute values in the T1, so as to form a discretization fault diagnosis information table T2; using a rough set discernable matrix and a nuclear theory to carry out attribute reduction and optimal feature selection on the T2, so as to form a reduction fault diagnosis information table T3; and using the T3 to establish the Bayesian network classifier, so as to realize efficient and quick intelligent fault diagnosis. The intelligent fault diagnosis method avoids the 'curse of dimensionality' problem existed in a Bayesian network diagnostic method, overcomes weaknesses of rigid reasoning and critical misjudgment in a rough set diagnostic method, and greatly improves the efficiency and accuracy of fault diagnosis.

A system and method for provisioning software on a plurality of computational nodes in a distributed computing environment. A plurality of support processing requirements associated with a software product is accepted. The plurality of requirements is expanded into multiple sets of installation requirements. At least one set of installation requirements in the multiple sets of installation requirements are minimized to produce at least one minimized set of installation requirements. A determination is made as to whether any pair of requirements in the minimized set of installation requirements includes a pair of conflicting requirements. A determination of whether the software product allows each requirement in the pair of conflicting requirements to be located on separate nodes is also made. At least one multi-node installation topology is determined for the software product. The multi-node installation topology includes a plurality of installation requirement sets for each node in the multi-node installation topology.

The invention belongs to the technical field of data mining in information technologies, and relates to an electrification detection data processing method based on a data mining technology, in particular to a state analysis technology and method for a transformer, a breaker, an arrester and others in electrical equipment. The processing method includes the steps that according to maintenance lag existing in traditional state maintenance of the electrical equipment and redundant work of overmuch planned maintenance, and a data mining state analyzing model with a rough set and a decision-making tree combined is built and combined with an existing electrical equipment state maintenance system to build the electrification detection data processing system based on the data mining technology. The rough set and the decision-making tree fusion technology is applied to state data analysis of the electric equipment, according to existing state judgment standards, state data of an over 35 KV transformer, an SF6 breaker and other electric equipment are preprocessed, Gini coefficient indexing, attribute reduction, threshold value selection and other steps are performed on attributes, state analysis is performed on the basis, meanwhile a work state table is formed, and a corresponding processing scheme is given.

A system, method, and computer program product for provisioning software on at least one node in a plurality of computational nodes in a distributed information processing system are disclosed. The method includes accepting a plurality of requirements associated with a software product. The plurality of requirements is expanded into multiple sets of installation requirements. At least one set of installation requirements in the multiple sets of installation requirements is minimized to produce at least one minimized set of installation requirements. At least one installation topology is determined using Rough Set Theory for the software product based on the at least one minimized set of installation requirements. The at least one installation topology is compared to a set of capabilities included on at least one computational node to determine a respective set of missing resources for the at least one computational node.

Disclosed is an apparatus and method for estimating a state of charge (SOC) in a battery, in which the battery SOC is estimated using a fusion type soft computing algorithm, thereby accurately estimating the battery SOC in a high C-rate environment. The apparatus includes a detector unit for detecting current, voltage and temperature of a battery cell; and soft computing unit for outputting a battery SOC estimation value of processing the current, the voltage and the temperature detected by the detector unit using a radial function based on a neural network algorithm. Especially, the soft computing unit combines the neural network algorithm with any one of a fuzzy algorithm, a genetic algorithm (GA), a cellular automata (CA) algorithm, an immune system algorithm, and a rough-set algorithm, and thereby adaptively updates the parameters of the neural network algorithm.

The invention discloses a fault diagnosis knowledge acquiring system, which comprises a semi-automatic knowledge acquiring module, an automatic knowledge acquiring module, an automatic knowledge base maintaining module, a knowledge base, a fault tree information base and a case base. The automatic knowledge acquiring module summarizes and concludes a new rule from a large number of cases stored in the case base according to the rough set theory and stores the acquired rule in the knowledge base, thus realizing automatic expansion of the knowledge base. The automatic knowledge base maintainingmodule realizes a function of automatically maintaining the knowledge base by using the character set closure and rule implication method, and the functional module can perform redundancy and circulation rule check on a rule base and provide the check result to the domain experts for judgment and processing. The invention acquires the known knowledge by using a fault tree, thereby not only indicating the logic relation in knowledge but also providing convenience for the user to maintain.

The invention discloses a facial expression recognition method based on a rough set and mixed features. The method includes the following steps that (1) facial detection is carried out; (2) local geometric distortion features are extracted by means of a method combining an active appearance model and the rough set; (3) overall features of expressions are extracted with the combination of an improved weighting primary component analysis method and the rough set; (4) feature fusion is carried out on the extracted local geometric distortion features and the overall features by means of kernel canonical correlation analysis under a high-dimensional small sample to eliminate feature redundancy, and fused typical features are obtained; (5) the fused typical features serve as observation vectors of a discrete HMM for classification and recognition, and a classification result is obtained. As is presented by experiments, the improved method can shorten facial expression recognition time and improve the facial expression recognition rate.

The invention discloses a fast positioning method for video events from rough state to fine state, comprising the following steps of: roughly matching an inquired event with an actual video to obtain the start and ending points for dividing the actual video time, extracting an interesting region of each frame of image in the actual video to finish space division of the actual video, performing space division and time division to the actual video to obtain a series of interesting time bodies, finely matching the interesting time bodies with the inquired event to construct a relative body, using a global significance test to identify whether each interesting time body has the event related to the inquired sample on the relative body, and finally, using a post-process method to control the time region positioning and displaying around the best matched obvious points. The method adopts the roughly matching method to remove large amount of irrelevant time regions to effectively reduce the matching and searching space so that the finely matching is only performed between the interesting time body and the inquired event, so the method improves the searching speed.

The present invention discloses a mechanical product design two-dimensional knowledge pushing method based on design intents. The mechanical product design two-dimensional knowledge pushing method mainly comprises: establishing a mechanical product design intent database, establishing a design intent acquisition and decomposition method, constructing a design intent attribute table by a rough set theory and performing reduction to obtain the simplest design intent set; establishing a mechanical product design knowledge ontology database; and calculating similarity degrees between intent elements and a compared knowledge ontology in a sequential traversing manner by using an improved similarity degree matching algorithm based on a knowledge ontology concept semantic distance and a concept attribute, sequentially carrying out matching on each intent element, of which the similarity degree is greater than a threshold value, in an intent element set from large to small according to the similarity degrees until completing all the matching, and completing matching of the knowledge ontologies by utilizing a text semantic similarity degree calculating method. The mechanical product design two-dimensional knowledge pushing method can solve the problem of low knowledge intelligent degree of knowledge pushing in the mechanical product design, can improve efficiency of mechanical product design and is suitable for the requirement of the manufacturing information engineering technology for deepening development.

The invention proposes a rough set based image segmentation method for quickly inhibiting fuzzy clustering. The method is used for solving the technical problems of low running speed, low segmentation accuracy and poor noise robustness of an existing image segmentation method. The method is implemented by the steps of 1, inputting a to-be-segmented image I1; 2, calculating a weighted mean of local information and a mean of non local information of pixel points xi in the image I1; 3, obtaining a reconstructed image; 4, clustering a grey level histogram of the reconstructed image; 5, judging whether a current iterative frequency is greater than a maximum iterative frequency T or not, and if yes, performing the step 6, otherwise, adding 1 to the iterative frequency and performing the step 6; 6, outputting a membership matrix and a clustering center of the obtained reconstructed image; and 7, obtaining segmented images. According to the method, the running speed of image segmentation is increased, the accuracy of segmentation is improved, and the noise robustness is enhanced; and the method can be used for feature extraction and target identification of artificially synthesized images, medical images and natural images.

The invention discloses a grid fault diagnosis method based on data mining. On the basis of the characteristics that grid measurement data quantity is large and dimensionality is high, grid fault diagnosis is achieved by means of a data mining technique. Firstly, the data correlation degree is evaluated by means of a neighborhood rough set method, reduction processing is carried out on data by means of a greedy search algorithm, then failure diagnoses are classified according to features by means of a competition coagulation clustering algorithm, and then failure diagnosis of a grid is achieved on the basis of a fuzzy association rule algorithm. According to the method, failure detection, classification and positioning of various devices and elements which break down in the grid with mass data can be carried out, ultimately, an IDEA integrated development tool and an MySQL database are used, an object-oriented program design concept is adopted, and a grid failure diagnosis system in a B/S mode is developed.

The invention relates to a gas-liquid two-phase flow pattern recognition method based on digital image processing technology, which is characterized in that the method uses a high-speed camera to gain the gas-liquid two-phase flow image in a horizontal pipeline under the different working conditions; the characteristics of invariant moment and gray level co-occurrence matrix of the image are extracted by the image processing technology; the characteristic fusion is implemented by using a rough set theory to reduce the characteristic dimensions, and the characteristic vector forms a flow pattern sample to implement the training for a support vector machine in order to complete the mapping from the characteristic space to the flow pattern space and finally realize the flow pattern recognition. The adopted rough set theory fuses image texture information and shape information, improves the recognition precision of a classifier, simultaneously reduces the training time, and can roundly reflect the characteristics of flow pattern image; the dependent degree and the generalization capacity of the flow pattern recognition method of the support vector machine for the sample data are better than the BP neural network; the invention has the shorter training time, and is applied to the flow pattern online recognition.

A system and a method for imputing missing values and a computer program product thereof are applicable to a data matrix. The system includes a storage unit having the data matrix and a computing device. The computing device finds complete and incomplete data transactions from the data matrix, finds at least one target data transaction approximate to each incomplete data transaction from the complete data transactions, and obtains known data at corresponding positions to compute an initial estimated data to replace unknown data. Then, a correction data transaction containing the initial estimated data is selected from the incomplete data transactions, a rough set of the selected initial estimated data is found in a manner of grouping same data into one group, and a numerical value correlated to the initial estimated data is found and used to compute an imputed data, so as to impute the imputed data into the original estimated data.

The invention discloses an image segmentation method based on genetic rough set C-mean clustering, which mainly solves the problem that the conventional method has poor robustness, easily falls into local optimum and loses too much local information. The method comprises the implementation steps of: (1) inputting a to-be-segmented image; (2) extracting image texture features; (3) generating clustering object data; (4) initializing population; (5) updating membership; (6) dividing the clustering object data; (7) updating the population; (8) calculating an individual fitness value; (9) evolving the population; (10) judging whether a termination condition is satisfied; (11) generating an optimal individual; (12) marking; (13) generating segmented images. In the method, the texture features of each pixel of the image are extracted, and the texture features are marked through the C-mean clustering method based on the genetic algorithm and the thought of rough set so as to divide the pixels, thus, stability of image segmentation is improved, and more accurate image segmentation result is obtained.

The invention provides a selection optimization method of a temperature measurement point combination for positioning errors of a numerically-controlled machine tool under thermal effect. The selection optimization method is capable of identifying the influence of the temperature measurement point in each position on the positioning errors of the machine tool based on a grey correlation policy and a rough set theory. The selection optimization method comprises the following steps: k temperature sensors are mounted in special positions of the machine tool to measure the real-time temperature values, changing over time, of the machine tool during operation, and meanwhile, a laser interferometer is used for measuring positioning error values affected by temperatures; n sensitive temperature measurement point positions are screened out by use of the grey correlation policy; the positioning errors and the temperature data of the machine tool are preprocessed according to the principle of the rough set theory and a policy table is established; m feasible temperature point combinations are obtained by use of rough set reduction software; the optimal temperature measurement point combination of the machine tool is identified by virtue of comprehensive analysis. The selection optimization method of the temperature measurement point combination for the positioning errors of the numerically-controlled machine tool under the thermal effect is capable of solving the problem of excessive temperature measurement points or poor compensation model robustness in the positioning error compensation modeling process of the numerically-controlled machine tool.

The invention discloses a video target multi-target tracking method and device and a storage medium. The video target multi-target tracking method comprises the following steps: obtaining a first image feature of a target object and a second image feature of an observation object; performing feature similarity calculation of N feature categories on the first image feature and the second image feature to obtain N groups of feature similarity results; screening the N groups of feature similarity results based on the feature similarity of the rough set, and fusing the screened results to obtain afeature fusion result; based on maximum entropy intuitionistic fuzzy clustering, performing association cost matrix calculation according to the feature fusion result to obtain an association cost matrix calculation result; judging whether the target object is associated with the observation object or not according to the association cost matrix calculation result, and if yes, updating a target model; and if not, performing target trajectory management on the target object.