186 results about "Retrieval algorithm" patented technology

A tag-mediated review system for electronic content that ranks content based upon several ascertainable factors, including the implicit engagement with the content by users (typically online viewers); the expertise and success in the topical domain by reviewers; and the review activity of users or viewers. The system extends implicit metrics to tracking of user interaction including the receipt of video and audio content over the web. Document ranking is mediated by tags that identify the subject matter addressed by an associated document as the ranking mediators for computing content quality scores values for user and reviewer activity. The tag-mediated content quality scores, which are content quality scores for the subject matter area identified by the associated tags, can be exposed to end-users and used directly or indirectly to complement search engine retrieval algorithms.

A confidentiality preserving system and method for performing a rank-ordered search and retrieval of contents of a data collection. The system includes at least one computer system including a search and retrieval algorithm using term frequency and / or similar features for rank-ordering selective contents of the data collection, and enabling secure retrieval of the selective contents based on the rank-order. The search and retrieval algorithm includes a baseline algorithm, a partially server oriented algorithm, and / or a fully server oriented algorithm. The partially and / or fully server oriented algorithms use homomorphic and / or order preserving encryption for enabling search capability from a user other than an owner of the contents of the data collection. The confidentiality preserving method includes using term frequency for rank-ordering selective contents of the data collection, and retrieving the selective contents based on the rank-order.

The invention discloses an image retrieval method, which is realized on the basis of a vocabulary tree level semantic model. Firstly, the characteristics of SIFT (scale-invariant feature transform) comprising color information of an image are extracted to construct the characteristic vocabulary tree of an image library, and a visual sense vocabulary describing image visual sense information is generated. Secondly, the Bayesian decision theory is utilized to realize the mapping of the visual sense vocabulary into semantic subject information on the basis of the generated visual sense vocabulary, a level semantic model is further constructed, and the semantic image retrieval algorithm based on content is completed on the basis of the model. Thirdly, according to relevant feedback of a user during a retrieval process, a positive image expandable image retrieval library can be added, and the high-level semantic mapping can be revised at the same time. Experimental results show that the retrieval method is stable in performance, and the retrieval effect is obviously promoted along with the increasing of feedback times.

A system and method to organize information on the internet for rapid and organized retrieval. Registrants of websites can register URLs by specifying the URL and associated descriptors. A bot automatically determines URLs and metadata associated with the registered URL. The URLs and descriptors and / or metadata form a URL database. Search terms entered by users can be indexed against a knowledge database using one or more retrieval algorithms to provide keyword associations. The knowledge database further includes a knowledge acquisition and retrieval system and method that include at least one first memory segment, and a distinct second memory segment, wherein elements of the at least one first memory segment reciprocally associate to elements of the second memory segment. Registrants can modify the knowledge database to incorporate non-traditional associations. The search term, keyword associations, and URL associations provide an organized search result that includes subcategories and cross-categories of information that can be further searched by the user. URL links can be provided in the search results.

A tag-mediated review system for electronic content that ranks content based upon several ascertainable factors, including the implicit engagement with the content by users (typically online viewers); the expertise and success in the topical domain by reviewers; and the review activity of users or viewers. The system extends implicit metrics to tracking of user interaction including the receipt of video and audio content over the web. Document ranking is mediated by tags that identify the subject matter addressed by an associated document as the ranking mediators for computing content quality scores values for user and reviewer activity. The tag-mediated content quality scores, which are content quality scores for the subject matter area identified by the associated tags, can be exposed to end-users and used directly or indirectly to complement search engine retrieval algorithms.

The invention discloses an image retrieval algorithm based on a free-hand sketch. According to the method, a metric learning algorithm is used for obtaining a mapping function, the free-hand sketch and a real image are mapped to the same metric space, an L1 normal form in the metric space can directly represent the 'semantic similarity' of the sketch in an original space and a real photo, and the problem of cross-domain matching of the free-hand sketch and the real photo is well solved. Specifically speaking, a positive and negative sample pair is used for training the convolutional neural network, feature vectors of the photo and the free-hand sketch input in the space and labeled to be similar are put together, the feature vectors of images labeled to be not similar are pushed away, and an integrated retrieval method is put out finally. Experiments show that the retrieval performance of the method is greatly improved in the large-scale free-hand sketch retrieval data set.

A system and method to organize information on the internet for rapid and organized retrieval. Registrants of websites can register URLs by specifying the URL and associated descriptors. A bot automatically determines URLs and metadata associated with the registered URL. The URLs and descriptors and / or metadata form a URL database. Search terms entered by users can be indexed against a knowledge database using one or more retrieval algorithms to provide keyword associations. The knowledge database further includes a knowledge acquisition and retrieval system and method that include at least one first memory segment, and a distinct second memory segment, wherein elements of the at least one first memory segment reciprocally associate to elements of the second memory segment. Registrants can modify the knowledge database to incorporate non-traditional associations. The search term, keyword associations, and URL associations provide an organized search result that includes subcategories and cross-categories of information that can be further searched by the user. URL links can be provided in the search results.

The invention discloses an efficient image retrieval method based on an improved SIFT (scale invariant feature transform) feature, relates to the field of image processing and computer vision and belongs to a content-based image retrieval method. The method comprises the steps of establishing a gauss difference scale space, taking scale space detection extreme points as feature points, calculating the direction of the feature points and gradient module values in the direction, generating a global feature vector of the image feature points through a geographical statistical method, conducting global feature vector similarity matching to finish rough retrieval, generating feature descriptors on the basis of rough retrieval, and conducting feature descriptor matching to finish accurate retrieval. The novel image retrieval method is superior to a traditional SIFT algorithm, and is more suitable for requirements of image infringement examination in digital prints than an existing retrieval algorithm.

The invention relates to a road edge detection system and method based on laser radar and fan-shaped space segmentation. The method comprises the steps of 1, a laser radar scanning the surrounding environment of a vehicle to obtain reflection point cloud information and convert the reflection point cloud information into a locally constructed three-dimensional coordinate system; 2, preprocessing the point cloud data, and separating and extracting ground data in each frame of point cloud; 3, dividing the space in the coordinate system into fan-shaped structural bodies according to the data characteristics of the laser radar and the point cloud, and identifying the road extension direction according to the ground information and the fan-shaped structural bodies; 4, extracting road edge candidate points in the point cloud by using a parallel road edge retrieval algorithm; 5, clustering the road edge candidate points, and eliminating an interference point set according to fan-shaped spatial features; and 6, performing B spline curve fitting on the finally determined road edge point to obtain a road edge detection result. The method is high in adaptability, capable of adapting to roadsof various shapes and reducing the influence of obstacles, high in precision and reduction degree, high in reliability and low in error rate.

The invention belongs to the technical field of mode recognition and information processing and provides a mass image retrieval system based on cluster compactness. Steps include 1, calculating local features of images in a sample image library and a test image library; 2, calculating cluster compactness of each image, namely clustering the local features to acquire each type of cluster centers, counting a local feature distribution histogram and spatial statistical information of each cluster, and generating cluster compactness; 3, randomly sampling cluster compactness of the sample image library, clustering components of the cluster centers in the sampled cluster compactness to generate a vocabulary tree, and quantizing the cluster compactness of the images in the test image library to the vocabulary tree to generate corresponding inverted files; 4, retrieving by a modified retrieval algorithm based on the vocabulary tree, namely retrieving, by retrieving the inverted files in the vocabulary tree and calculating the weight of similarity between retrieval images and the image library image cluster compactness.

The invention discloses a cross-modal retrieval algorithm based on mixed hypergraph learning in subspace, based on the cross-modal public subspace learning of canonical correlation analysis. The similarities inside and among modals are calculated by a mapping of the public subspace. A mixed relation matrix is calculated by the similarities inside and among the modals. A mixed hypergraph model is built by the extraction of the relation matrix. And the cross-modal retrieval and sample sequencing are conducted by adoption of hypergraph learning finally. With aiming at cross-modal heterogeneous variations and high order relation among samples, the algorithm instance applies the hypergraph model combined with the cross-modal public subspace learning to the cross-modal retrieval, so that the model is capable of considering the similarity among the modals and the similarity inside the modals simultaneously and giving consideration to the high order relation among a plurality of the samples meanwhile, improving the final precision ratio and the final recall ratio of the cross-modal retrieval. The algorithm is capable of effectively improving the performance of the cross-modal retrieval and greatly enhancing the precision ratio and the recall ratio of the cross-modal retrieval.

The invention discloses a multi-dimensional inverted index and quick retrieval algorithm for large-scale image visual features. The algorithm includes: training a multi-layer codebook required for enhanced residual quantification by visual features of images and building a multi-dimensional inverted index with the trained codebook; according to the trained codebook, quantifying and encoding the image visual features, and according to calculated codes, inserting the codes into corresponding inverted lists in the inverted index; querying the image visual features to query the built multi-dimensional inverted index so as to obtain a query candidate set; optimizing the query candidate set by adaptive hypersphere filtering, and ranking filtered query results so as to finish retrieval of the image visual features. The algorithm has the advantages that as the image features are quantified and encoded, efficiency of quantifying the image features is improved; the multi-dimensional inverted index is built with the generated image codes, the inverted index can be built just by training few codebooks, and speed of building the index structure is increased.

An imaging spectrometer, covering the visible through infrared wavelengths, which disperses the light by a plane diffraction grating behind a wedged optical element. This design uses an achromatic doublet lens with a reflective coating on its convex back surface to produce the spectra on a flat detector. Spatial keystone distortion and spectral smile are controlled to less than one tenth of a pixel over the full wavelength range, facilitating the use of simple retrieval algorithms.

The invention discloses a backscatter-and-oblique-ionograms-based joint inversion method for ionospheric parameters. The method comprises the following steps of: determining an ionospheric model required by the backscatter detection and oblique detection joint inversion as a quasi-parabolic model; setting parameters of the ionospheric model, and calculating by using the set ionospheric parameters to obtain a simulated backscatter ionogram front edge and a simulated oblique ionogram; adding random errors into the simulated backscatter ionogram front edge and the simulated oblique ionogram, wherein the simulated backscatter ionogram front edge and the simulated oblique ionogram into which the random errors are added serve as the backscatter ionogram front edge and the oblique ionogram for the final inversion; and performing data sampling on the backscatter ionogram front edge and the oblique ionogram which are used for the final inversion and comparing the difference between the inversed result and the set parameters of the ionospheric model to judge the effectiveness of a retrieval algorithm, wherein the sampled data serves as the input of the inversion. By introducing the novel ionospheric joint detection technology and taking detection data, in different forms, of the same detection area as the input of the retrieval algorithm, the precision and stability of the inversion are effectively improved.

The present invention provides a new approach for processing hyper-spectral radiance data. It uses a transformation matrix to convert an instrument radiance spectrum into a pseudo-monochromatic radiance spectrum. The pseudo-monochromatic radiance spectrum is produced by an empirical transform of the instrument channel spectrum to a monochromatic equivalent spectrum (i.e., a pseudo-monochromatic spectrum). Eigenvector regression is used to produce the empirical transformation. Although the transformation does not produce the monochromatic radiance spectrum without error, the transformation error is generally well below nominal instrument noise levels for most spectral channels. The reduction in instrument noise results from a noise filtering effect of the eigenvector transformation. One of the advantages of the present invention is that it eliminates the need to build different fast radiative transfer models (RTMs) for different observing instruments, since the retrieval of geophysical parameters is based on an inversion of the monochromatic radiative transfer model. Although a different transformation matrix is required for different instrument spectral channel characteristics, the production of this transformation matrix is straightforward and simpler than the production of an accurate channel radiance fast model.

The invention provides an elastic medium wave equation return wave travel time retrieval method. The method comprises the steps of designing a new travel time residual target function; deducing an adjoint equation and a background parameter gradient formula under the new target function; calculating the gradient of background parameters; processing the gradient by using the quasi-Newton method retrieval algorithm; calculating the iteration step size by using the parabola fitting method; updating a background parameter model until the convergence condition is met. The method has the advantage that the precision of elastic medium return waveform retrieval is improved by using the new travel-time target function, so that reliable longitudinal wave velocity and transverse wave velocity initialmodels can be obtained.

The invention discloses a question matching method, a terminal and a computer readable storage medium. The method comprises the following steps that: receiving the questions of a user, and inputting the received questions of the user to a knowledge base; in the knowledge base, adopting a retrieval algorithm to retrieve each prestored question to retrieve a plurality of questions matched with the questions of the user; adopting a classification algorithm to determine the question with a highest matching degree with the question of the user in a plurality of matched questions; and searching theanswer of the question with the highest matching degree in the knowledge base, and feeding back the searched answer to a display interface. By use of the method, question matching accuracy is improved.

The invention specifically discloses a condition diagnosis system for transformer fault types, belonging to the technical field of condition monitoring and fault diagnosis of power equipment. The condition diagnosis system comprises an oil sample collection module, an oil-gas separation module, a gas separation module, a gas detection module, a data processing module and a data transmission module. The condition diagnosis system provided by the invention successfully realizes rapid, accurate and real-time detection of the composition and concentrations of dissolved gases in the insulating oil of a transformer, avoids the shortcoming of lack of monitoring means with long-term stability and credible precision in the operation and maintenance of the transformer, and provides reliable data for online evaluation and analysis of the operating state and remaining life of the transformer; and a sample retrieval algorithm based on Euclidean distance is innovatively introduced for identification of fault types, so the condition diagnosis system overcomes problems in fuzziness and applicability of data about dissolved gases in the insulating oil, is improved in accuracy and timeliness of fault recognition, and achieves the goal of comprehensive optimization of risks, efficiency and cost in power safety production.

The invention provides an archive management system based on blue light storage and an ant colony search algorithm thereof, comprising an archive collection system, an archive storage server, an optical disk library management server and an optical disk library. Specifically, electronic archive files are assembled, compressed and indexed by the archive collection system in the form of data streams. The ant colony search algorithm has the following specific steps: 1. Constructing a file retrieval path; 2. Updating that pheromone of archives retrieval; 3. Optimizing evaporation factor of searchpath. The algorithm can automatically adjust the information volatility factor, analyze the keywords, and call the relevant persistence layer data through the algorithm. The archives management systemadopts the optimized retrieval arithmetic, According to the characteristics of archives, ACA algorithm is used, its process and algorithm are analyzed, by combining the pheromone volatility, the effect of the pheromone volatility factor on the efficiency of algorithm retrieval is improved, and by combining differential evolution processing, efficient retrieval efficiency of massive data is realized.

The invention discloses a method and a device for retrieving the similarity of pictures and belongs to the field of the graphic images. The method comprises the steps of obtaining the to-be-retrieved picture characteristics of to-be-retrieved pictures, hashing the to-be-retrieved picture characteristics by use of an LSH (Locality Sensitive Hashing) algorithm to generate hash values of the to-be-retrieved pictures, finding database picture hash values of which are matched and similar to the hash values of the to-be-retrieved pictures om a hash table corresponding to every hash values of the to-be-retrieved pictures, finding database pictures according to the database picture hash values, and according to Euclidean distances of the characteristics of the database pictures and the characteristics of the to-be-retrieved pictures, selecting a preset number of database pictures from the database pictures having relatively smaller Euclidean distances. The method and the device have the characteristics that the problem that a user cannot retrieve the most similar pictures from a plurality of results obtained by the LSH algorithm at the present is solved, and the advantages of LSH in reducing the time-space complexity in picture similarity retrieval and supporting high-dimensional data retrieval are better developed by virtue of the combination of the LSH algorithm and a linear retrieval algorithm.

An image index algorithm based on information abrupt includes converting color data of sample image inputted by user from GRB space to be HSV space and making normalized treatment on it then carrying out initial division on image according to pixel correlation degree and information abrupt, carrying out circulating combination as per similarities of divided pixel blocks, dividing image to be 3x3 subblock and extracting out feature vectors of nine regions separately, selecting interested region from nine regions by user and comparing it with image to be indexed on their proceed similarities to obtain index result of proceed image.

The invention discloses a statement sorting method and system applied to an answer fusion system. The method comprises the following steps that: preprocessing input target data to obtain a statement pair of an index form, wherein the statement pair is a candidate answer; carrying out feature extraction on the statement pair to obtain the relationship probability of every two adjacent sentences inthe statement pair; according to the relationship probability and a solution space retrieval algorithm, obtaining a sorting result, wherein a target function solved by the retrieval algorithm is Score(s,o)= [Sigma]<k>i=1[Sigma]<k>j=i+1 score (s,o,i,j); and according to the sorting result, obtaining a target answer. The step of carrying out feature extraction on the statement pair to obtain the relationship probability of every two adjacent sentences in the statement pair comprises the following steps that: judging whether sentences in the statement pair are in a sequential order relationshipor not, utilizing a seq2seq model based on an attention mechanism to carry out feature extraction on the statement pair, and inputting an extracted feature into a feedforward neural network to obtaina sentence sequential order probability. By use of the method, sentences extracted from different answers can be sorted, so that the obtained target answer is smooth, is favorable for understanding and is high in readability, and user experience is improved.

The invention discloses a method for providing network service resources. The method comprises the following steps: firstly, classifying the network service resources; then, providing a retrieval scheme according to the interest of a user, and providing the classified network service resources according to the retrieval scheme; firstly extracting feature vectors of to-be-classified service resources when the network service resources are classified, then, computing the probability of each attribute in each feature vector in each category and the weight of each attribute, then, acquiring the probability that each attribute belongs to each category by utilizing a weighting Naive Bayes formula, and selecting the maximum as the classification category of each service resource. According to the method disclosed by the invention, when the service resources are classified, by utilizing the bayesian classification algorithm and combining the calculation of attribute similarity, the classification accuracy of the service resources is improved, and when the service networks are provided for the user, a user interest-based personalized service resource retrieval is explored under the framework of a vector space model, so that the complexity of time and space in the retrieval algorithm is reduced, and the retrieval efficiency is improved.

The invention provides a high-dimension vector rapid searching algorithm based on a block distance and belongs to the field of data processing such as multimedia information searching, intelligent information processing, data mining, and the like. In the invention, an index structure Block B-tree which is converted from high dimension to one dimension and is based on the block distance is provided; a high-dimension vector is mapped into one-dimensional key values by adopting the block distance of the high-dimension vector to a reference point; and the index structure B+-tree is used for managing the key values, and each key value of a leaf node layer is bound with a pointer pointing to a corresponding high-dimension vector. During searching, the same mapping method is used for mapping a query vector into one-dimension query key values, and then similarity calculation is only performed on the high-dimension characteristics of the key values close to the query key values, thereby reducing the calculated quantity and greatly increasing the searching speed. In a similarity matching algorithm of the high-dimension vector, the block distance is a frequently-used measurement way, the operation of the algorithm is simple, and the searching efficiency is higher, while most of the current index structures are provided based on Euclidean distance matching measurement. The index structureprovided by the invention not only supports searching based on the Euclidean distance matching way but also directly supports searching based on the block distance measurement way.

The invention provides a retrieval algorithm for deleting repetitive data in cloud storage. When deletion of the repetitive data needs to be carried out, a file similar to a newly written file in a whole storage system can be retrieved at first; furthermore, a threshold value is reached; then, the two files are precisely compared; the repetitive data is abandoned; and only different data and index information are reserved. A certain amount of file fingerprint data is randomly extracted in a sampling manner; a sampling method and the sampling amount are considered; in combination with the sample similarity, a file repetitive rate function is constructed; a redundant file is abandoned by setting a repetitive rate threshold value; therefore, deletion of repetitive files is realized; the storage space is saved; the method has the characteristics of being rapid in calculation speed and high in deletion rate; and thus, the retrieval algorithm is more suitable for large data and cloud storage environments.

A system is provided for integrative analysis of intrinsic and extrinsic audiovisual information, such as a system for analysis and correlation of features in a film with features not present in the film but available through the Internet. The system comprises an intrinsic content analyser communicatively connected to an audio-visual source, e.g. a film source, for searching the film for intrinsic data and extracting the intrinsic data using an extraction algorithm. Further, the system comprises an extrinsic content analyser communicatively connected to an extrinsic information source, such as a film screenplay available through the Internet, for searching the extrinsic information source and retrieving extrinsic data using a retrieval algorithm. The intrinsic data and the extrinsic data are correlated in a multisource data structure. The multisource data structure being transformed into high-level information structure which is presented to a user of the system. The user may browse the high-level information structure for such information as the actor identification in a film.

The invention discloses a multi-outlet routing selection method, which comprises the following steps of performing a path detection module according to each network outlet server to realize real-timedetection of a destination address, uploading the destination address to a path management center, judging a routing strategy by the path management center, and determining an optimal path of the samedestination address in a plurality of network outlets. A user-defined module of a netfilter adopts an interval tree retrieval algorithm; therefore, the destination addresses of all the data packets are quickly classified and marked (a network outlet is specified), the user permission level is dynamically matched according to the source addresses of the data packets, and the corresponding available network outlet is specified based on the user permission level and the identification code of the network outlet server. Routing policy routing realizes routing outlet selection according to the fwmarkb packet identifier, and the method can adapt to dynamic adjustment of multiple outlets of the network, and realizes high-precision dynamic automatic selection of the multi-outlet routing.

The invention discloses an urban rainfall inversion algorithm based on microwave attenuation characteristic response fingerprint identification, includes the acquisition and preliminary processing ofmicrowave signals, the separation and classification of clutter signals, the establishment of clutter characteristic response fingerprint database, the identification and removal of clutter, rainfallinversion, scale matching of multi-source rainfall data, the optimization of model parameters and the real-time correction of inversion results. The invention can quickly judge and remove the influence of different types of clutter in the microwave attenuation signal under the urban environment, and provides urban multi-resolution rainfall big data with a spatial resolution of 100 meters-200 meters and a time resolution of 5 minutes-10 minutes.

The invention discloses a quantitative description method for an assembly model during model retrieval, which is used for solving the technical problem of low retrieval efficiency of a traditional quantitative description method for an assembly model. In the technical scheme, the quantitative description method comprises the following steps calculating a Euclidean distance between two arbitrary points on the surface of a part model; extracting a distance distribution histogram; expressing each part as a point in a 1024-dimensional space; carrying out separate and accurate quantitative description on the part model by using such point; expressing the assembly model as a set with a series of points, wherein each point expresses a part in the assembly model; and carrying out quantitative description on the assembly model in a point set mode. By the quantitative description method, the description accuracy is ensured, and meanwhile, a description result of the part model is only dependent on the model per se and is not affected by other factors; and compared with the background art, the quantitative description method has the advantages that the point set mode is used for quantitatively describing the assembly model, the complexity of a retrieval algorithm is reduced, and the retrieval efficiency is improved.

The invention belongs to the technical field of image and video processing, and specifically relates to a deep learning-based approximate video copy retrieval algorithm. The algorithm is divided intotwo stages: an offline index construction stage and an online retrieval stage. In the offline index construction stage, feature values of sampling frames are extracted by using a deep convolutional neural network, and then indexes are established for all the video sampling frame sets in a video library by adoption of a k-d tree; and in the online retrieval stage, feature values of sampling framesof a queried video are extracted by adoption of the same method, similar candidate videos are searched from an index library by using an approximate nearest neighbor search method, similarities between all the candidate videos and the queried video are calculated finally, and the similarities are sorted from high to low to obtain an approximate copy retrieval result. According to the algorithm, the whole retrieval process can be greatly accelerated, and similarities between candidate videos and queried videos can be obtained for subsequent steps at the same time, so that the retrieval speed isfurther improved.