47results about How to "Effective portrayal" patented technology

The invention discloses a method for estimating a state during a running process of an automobile. The method comprises the following steps: analyzing a driving feature of the automobile and establishing a motion model of the automobile; introducing a non-integrity constraint for establishing a velocity error estimating equation; estimating errors along side direction and vertical direction during the running process of the automobile on the basis of a velocity and three attitude angles of the automobile; and establishing a slip angle estimating equation of the automobile. The stability of the automobile during the running process can be judged according to the velocity error estimation and slip angle estimation. According to the method, the on-line detection for a front-wheel steering angle of a four-wheel automobile can be realized by utilizing the motion model of the automobile. By adopting the method for estimating the state during the running process of the automobile, the real-time calculation and monitoring for state parameters of the automobile during the running process can be realized, for supplying data reference to safety running of the automobile.

The invention discloses a remote sensing image variation detecting method on the basis of weighted Gabor wavelet characteristics and two-stage clusters. The processed objects include optical remote sensing images and SAR (synthetic aperture radar) images and the remote sensing image variation detecting method includes (1) generating difference images according to remote sensing image types; (2) subjecting the difference images to Gabor wavelet transform; (3) extracting multiscale and multidirectional characteristics of the Gabor wavelet transform of the difference images; (4) designing the weighting coefficient and acquiring the weighted Gabor wavelet characteristics; (5) clustering the weight Gabor wavelet characteristics by means of the two-stage cluster strategy; (6) acquiring variation detecting results. By the remote sensing image variation detecting method, loss of marginal information is reduced, stronger, weak and slight varying areas can be detected at the same time, the total mistake pixel number is decreased, more detail information is reserved and variation results can be effectively extracted.

The invention provides a sensitive image recognizing method in interaction of inner and outer power networks. The method includes the steps that (1) sample images are collected in an inner power network system to form a sample image set, and sensitive and insensitive marking is carried out on the sample images in a manual marking mode; (2) feature items of the sample images are selected; (3) a feature data set of the sample image set is extracted according to the selected feature items; (4) a machine learning method is adopted, and a classification model is obtained through training according to the feature data set of the sample image set and the corresponding sensitive or insensitive marks; (5) on the basis of the classification model, sensitive image recognizing is carried out; when the misjudgment rate is smaller than a set misjudgment threshold value, the result that the current classification model meets requirements of an expected target can be determined, and training is finished; when the misjudgment rate is larger than or equal to the misjudgment threshold value, the feature items of the sample images are selected again, and the step (3) is carried out. According to the method, sensitive images are recognized based on the machine learning method, and the excellent classification model can be obtained under the condition that the number of the samples is limited.

The invention discloses an energy-saving risk probability considered provincial power grid electricity purchasing decision making system. The system comprises a power plant tradable electricity quantity input unit, a random risk probability environment simulation unit, an electricity purchasing scheme generating unit and an electricity purchasing cost and energy-saving risk probability output unit. An electricity purchasing model is built inside the system, and a provincial power grid electricity purchasing scheme is obtained. According to the energy-saving risk probability considered provincial power grid electricity purchasing decision making system, the network loss factor of market transactions in an energy-saving electricity market is considered, and unit transaction electricity quantity energy-saving benefit evaluation indexes are built; with the reference of the probability risk concept, the market price, hydropower unit electricity quantity generation and randomness of system load electricity quantity requirements are considered, power grid company market electricity purchasing energy-saving risk probability indexes are built, an energy-saving risk probability considered provincial power grid electricity purchasing model is built in a randomly-planned theoretical framework for achieving optimization management of power grid company market electricity purchasing energy-saving risks, and a genetic algorithm with the dominant embedded target and the Monte-Carlo random simulation technology is adopted for solution.

The invention relates to the carotid ultrasound image processing field and provides a reliable and automatic carotid ultrasound image measurement method. In the method, a mode of using a computer to assist the measurement is adopted so as to avoid a defect existing in a manual measurement mode. In the technical scheme, the carotid ultrasound image intima media thickness measurement method comprises the following steps of 1) carrying out image cutting, cutting and removing irrelevant information and only keeping a blood vessel portion; 2) automatically extracting a region of interest ROI (region of interest); 3) carrying out image filtering and filtering a plaque noise brought by echo reflection; 4) carrying out horizontal set segmentation; 5) initializing a mark field; 6) carrying out resegmentation; 7) carrying out postprocessing, postprocessing a segmented image, removing an error-dividing small area, acquiring a finally-segmented image, LII and MAI, and measuring intima media thickness. The method and the system are mainly applied to carotid ultrasound image processing.

The invention discloses a partitioning compressive sensing reconstruction method based on image block clustering and sparse dictionary learning, and belongs to the technical field of image processing. The method comprises the following steps that an image is read in, and is divided into sub-image blocks; compressive sampling is carried out on the sub-image blocks to achieve measurement; edge images in (K-1) directions are generated, PCA transformation is carried out on the edge images to generate (K-1) PCA bases, and then, a DCT base is taken for forming a union dictionary of K initial direction bases; the typical correlation coefficients between the measurement and the direction bases are calculated, and the sub-images are clustered to form K classes; the sub-image blocks in the K classes are reconstructed through a multivariable tracking algorithm; the reconstructed sub-image blocks are used for updating the K direction bases; whether the maximum number of times of iteration reconstruction is reached or not is judged; the reconstructed sub-image blocks are spliced together to obtain a reconstructed image of the original image; the image is output. According to the partitioning compressive sensing reconstruction method based on image block clustering and sparse dictionary learning, the blocking effect in the reconstructed image can be obviously weakened or removed in two reconstruction modes, and the method has a reconstruction effect on a natural image.

The invention relates to an image feature extraction method for garment image retrieval, belonging to the technical field of image retrieval. Firstly, a key region generation network based on key points and a key region fusion network for fusing global features and key region features of garment images are designed creatively. Then the key region generation network and the key region fusion network are added to the open source depth learning model VGG16 to obtain the depth learning model for clothing image retrieval. Then the model is converged by cross-training the key regions to generate thenetwork and the key regions to converge the network. Finally, the high-level features of the depth learning model proposed by the invention are extracted for the garment image retrieval task. The depth feature extraction method of the garment image provided by the invention can effectively improve the accuracy of garment image retrieval, and the method is simple and easy to realize.

The invention discloses a D2D information propagation modeling method based on user attributes and content attributes, belonging to the mobile communication field. The method comprises the following steps: 1, according to user history interaction, combining interaction factor and stability factor to quantify the social relationship among users; 2, establishing a user infection probability functionaccording to that social relations among users and the degree of interest of the us in the information content; 3, using the degree of user and neighbor user to represent user influence, and introducing the influence attenuation model to establish the user recovery probability function; 4, in combination with the SIR model, determining the D2D information propagation model base on the discrete-time Markov chain; 5, carrying out stability analysis of information propagation model based on user attributes and content attributes. The D2D information propagation model proposed by the invention can effectively improve the information propagation range and has higher matching degree with the actual network.

The invention discloses an edge-preserving and structure-preserving recovery method for a variation PDE (Partial Differential Equation) image. A structural feature detection function is introduced to a model to punish different functional spaces in the model so as to make the model get adaptivity, even obtain a high-quality recovery image. In the model algorithm implementation, a problem is converted into a plurality of easily solved sub-problems, and a numerical value model has a fast Fourier transform feature by adopting a periodic differential format to solve efficiently and stably. According to the method provided by the invention, a software package lays particular emphasis on converting the model into the quickly solved sub-problems, and the sub-problems have separability and use fast Fourier transform by adopting the periodic differential format, therefore the problem can be efficiently and stably solved and have a parallelism.

The invention discloses an inference method facing to indefinite context of general fit calculation, which uses set pair analysis technology and related coefficient concept to infer and integrate primary context data sensed by a bottom sensor so as to acquire a high-layer context information technical proposal in a general fit calculation environment, and uses the set pair analysis technology and vector correlation coefficients to achieve quick, convenient, intuitive and reliable inference for the indefinite context; and the set pair analysis method effectively reflects the opposite unity relation of definite-indefinite systems, accords with natural dialectics and human thinking mode, and has the significance of methodology. In various actual context sensing applications, the set pair analysis method is simpler, more convenient and more effective to describe indefiniteness than the methods using probability theory, fuzzy set and the like. A connection degree expression is written to a form of vector for short as mu (W) = (a, b, c), and is named as identity-diversity reverse vectors of the set pair. The cosine of an included angle between the identity-diversity reverse vectors reflects the closing degree of the identity-diversity reverse vectors, when the cosine value of the included angle is larger, the difference is smaller, and the similarity is larger.

The invention relates to an information fusion-based hidden Markov salient region detection method. The method of the invention can be applied to target region detection and recognition in a complex background. The information fusion-based hidden Markov salient region detection method is characterized by comprising the following steps that: firstly, three color contrast channels of an image are calculated; secondly, final tag images of the channels are calculated through maximum posteriori probability iterative calculation; salient images of the channels can be obtained through utilizing a tag class with maximum mutual information; and finally, information quantity of communicated regions in the three channels, which is larger than redundant information, is added together and is normalized, so that a final saliency image can be obtained. According to the information fusion-based hidden Markov salient region detection method of the invention, prior information of the image is fully utilized with data details preserved, and therefore, problems brought by region color information missing can be solved, and a wide salient region can be quickly focused without pre-training required.

The invention relates to a black-smoke vehicle detection method based on a multi-scale block LBP and a hidden Markov model. The method includes: (1), detecting a moving target from a monitoring videoand determining a key region; (2), extracting multi-scale block LBP features increasing scale and position information; (3) depicting dynamic features of the key region by using a hidden Markov modeland dividing current frames into black smoke frames and non-black smoke frames; and (4), carrying out analysis on a video sequence and identifying a black smoke vehicle further by combining a distribution characteristic of the black smoke frames. Therefore, automatic detection of black smoke vehicles is realized and the detection efficiency is improved; and because the dynamic features of the keyregion are depicted by using the hidden Markov model, the false alarm rate is reduced.

The invention relates to the technical field of rock fracture detection, and provides a rock fracturing fracture network topological structure depicting system, which comprises a detection shell. A memory, a processor and a computer program which is stored in the memory and can run on the processor are arranged in the detection shell. When the processor executes a computer program, the following steps are realized: acquiring rock in-place images at different deformation moments by adopting image equipment, and acquiring in-place sub-pixel crack spatial distribution based on each image; by adopting the technical scheme, image boundary conditions do not need to be simplified, different results caused by different structural discretization forms are avoided, and inherent crack information isnot contained; the rock fracture network can be effectively depicted, and the whole rock fracture network expansion process is obtained.

The invention relates to an information extraction method and device, electronic equipment and a storage medium, relates to the technical field of information processing, and aims to solve the problems that object meta-information is complex, lack of pertinence and cannot be accurately and efficiently applied in the related art, and the method comprises the following steps: obtaining a label set corresponding to a target object and account information of a target account, wherein the label set comprises at least two labels used for describing objects; based on a preset mapping mode, obtainingattention features obtained through mapping according to the account information, wherein the attention features comprise weights corresponding to all tags in the tag set, and the preset mapping modeis determined according to historical behaviors of sample objects corresponding to the sample accounts; and determining target object meta-information corresponding to the target account according tothe attention features and the label set. According to the embodiment of the invention, the attention characteristics associated with the user can be obtained, the attention of the user to each labelin the label set can be represented, and the target object can be described in a targeted manner.

The invention relates to the technical field of rock fracture detection, and provides a rock fracturing fracture network topological structure depicting method which comprises the following steps: acquiring rock in-place images at different deformation moments by adopting image equipment; and acquiring in-place sub-pixel crack space distribution based on each image; by adopting the technical scheme, image boundary conditions do not need to be simplified, different results caused by different structural discretization forms are avoided, and inherent crack information is not contained; the rockfracture network can be effectively depicted, and the whole rock fracture network expansion process is obtained.

The invention relates to a recommendation method and device. The recommendation method comprises: obtaining to-be-recommended data, wherein the to-be-recommended data comprises multiple pieces of user data, multiple pieces of project data, a user knowledge graph constructed based on the multiple pieces of user data and a project knowledge graph constructed based on the multiple pieces of project data; inputting the to-be-recommended data into a trained recommendation model to obtain a preference score prediction value of each user data to each item data in the to-be-recommended data; and obtaining at least one item data for each user data as recommended item data based on the preference score predicted value of each user data for each item data in the to-be-recommended data. According to the recommendation method and device disclosed by the invention, the TransR model is adopted as a spatial conversion network in the recommendation model, correct knowledge expressed in a triple form is effectively expressed, and a more efficient fusion mode of collaborative information and knowledge propagation is realized.

The invention provides a representation method and device for reservoir lithofacies. The method comprises the steps that according to distribution information of wells, a configuration unit and a lithofacies unit in a to-be-analyzed region, a configuration unit comparison profile and a lithofacies unit comparison profile are established; according to the configuration unit comparison profile, thelithofacies unit comparison profile and coordinate information of all the wells, a three-dimensional configuration unit comparison profile and a three-dimensional lithofacies unit comparison profile are constructed; a configuration unit boundary control line and a lithofacies unit boundary control line are picked, and a configuration unit interface and a lithofacies unit interface are established;according to the configuration unit interface and the three-dimensional configuration unit comparison profile, a three-dimensional configuration unit geologic model is established; and according to the lithofacies unit interface and the three-dimensional configuration unit geologic model, a three-dimensional lithofacies unit geologic model is constructed. By use of all embodiments, three-dimensional distribution characteristics of the configuration unit and the lithofacies unit can be accurately depicted, and the accuracy of geologic structure analysis is improved.

The invention provides a reference place recommendation method and device, equipment and a storage medium. The method comprises the steps: obtaining a reference place obtaining request initiated by atarget user terminal, the reference place obtaining request carrying the current position of the target user terminal; determining a plurality of target reference places within a preset range from thecurrent position; inputting the identifier of the target area to which the current position belongs and the identifiers of the plurality of target reference places into a pre-trained vector generation model to obtain vector representation of the target area to which the current position belongs and vector representation of the plurality of target reference places, wherein the vector generation model is obtained by training a preset model by taking an association relationship between regions of various sizes and each reference place as a training sample; determining a to-be-recommended reference place recommended to the target user terminal according to the similarity between the vector representation of the target area to which the current position belongs and the vector representation ofeach of the plurality of target reference places.

The invention discloses an MR image placenta segmentation method for a multi-task generative adversarial model. The method comprises the steps: constructing a training set and a test set, constructing a segmentation model through employing a generative adversarial network, training the segmentation model through employing the training set, the test set, and a total adversarial loss function of the segmentation model, and obtaining a trained segmentation model, finally, the placenta MR image is segmented through the segmentation model, segmented pictures of the placenta MR image are obtained, and a total adversarial loss function of the segmentation model is constructed based on maximization of discrimination loss and minimization of multi-task generation loss; the method has the advantages of being high in segmentation precision and capable of meeting clinical requirements.