658results about How to "Improve retrieval speed" patented technology

The invention discloses a multi-task layered image retrieval method based on a depth self-coding convolution neural network. The method is characterized by mainly comprising a multi-task end-to-end convolution neural network for deep learning and training recognition, a rapid visual segmentation detection and positioning method of a region-of-interest secondary screening module based on an RPN network, a coarse search of a full-graph sparse hash code, an area sensing semantic feature and matrix h accurate comparison and search based on the maximum response, and a region-of-interest selectivitycomparison algorithm. According to the method, the end-to-end training can be achieved, the interest region with higher quality can be automatically selected, the automation degree and the intelligent level of search by images can be effectively improved, and the image retrieval requirements of the big data age can be met by using little storage space at a high search speed.

The present invention relates to a fine-granularity files and directory edition management method based on snapshots, belonging to the multi-version document system field. The present invention separates a name space consisting of files and dirnames in a whole file system from an edition space representing the generating periods of different editions, and adopts relatively independent strategies to execute management, forming a hierarchical two-dimensional structure, i.e. forming a hierarchical structure from a root directory to a file in the name space; in the edition space, the editions of files and directory are organized through an index structure chronologically, forming a hierarchical structure in the edition space. The retrieval of the name space adopts an index strategy based on dynamic hash. The retrieval of the edition space adopts an index strategy based on a red-black tree. The directory edition and file edition respectively adopt a red-black tree structure variant aiming at the respective characteristics. The present invention can greatly improve the usability and the performance of the system, and controls the amount of consumption of time and space resulting from the maintenance of historical editions in an acceptable scope.

The invention relates to a visual target retrieval method and system based on target detection. The method comprises the steps that an IDF weighted cross entropy loss function is adopted to train a public target detection dataset, and a preliminary target detection model is generated; a retrieval dataset containing a target type designated by a user is adopted to slightly adjust the preliminary target detection model, and a final target detection model is generated; and feature extraction is performed on a visual target in a to-be-retrieved picture through the final target detection model, multiple convolution feature graphs of the to-be-retrieved picture are generated, the convolution feature graphs are aggregated through a spatial attention matrix, aggregate feature vectors are generated, and a picture matched with the aggregate feature vectors is retrieved in a picture library. According to the method, visual target retrieval and detection are associated, so that a candidate window prediction step is avoided; and the attention matrix is obtained by selectively accumulating the feature graphs, local descriptors of a convolution layer are aggregated into a global feature expression in a weighted mode, the global feature expression is used for visual target retrieval, and retrieval speed and precision are improved.

The invention belongs to the technical field of image retrieval and discloses an efficient cryptograph image retrieval method capable of supporting privacy protection under a cloud environment. The efficient cryptograph image retrieval method capable of supporting the privacy protection under the cloud environment comprises the steps that an image owner firstly extracts characteristic vectors of an image in a database and encrypts the image by an AES or RSA encryption manner; the owner establishes indexes using a hierarchical K-means algorithm based on the characteristic vectors and uses a safe mode Hash technology to encrypt the indexes; and the encrypted image and indexes are uploaded to a cloud server, and a retrieval operation is executed. According to the invention, the privacy protection of the cryptograph image retrieval can be achieved, and retrieval accuracy equivalent with public image retrieval can be reached. In addition, detailed safety analysis of the disclosed method is conducted, and experimental assessment of efficiency and accuracy is conducted in different datasets. Results indicate that the method disclosed by the invention can reach expected safety targets, and can increase retrieval efficiency on the premise of ensuring retrieval accuracy.

The invention discloses an image recognition method and device and a video monitoring device. The method comprises the following steps: detecting a target object in image information by using a detection model trained based on a depth learning algorithm; recognizing the feature information of the target object by the recognition model based on the training of depth learning algorithm; extracting the structural features of the target object from the feature information. Thus, by integrating the depth learning technology of artificial intelligence and the video surveillance technology in depth,using the detection model and recognition module trained by depth learning algorithm to detect the target object in the image information and recognize the feature information of the target object, and then extracting the fine structural features of the target object, the image recognition efficiency and recognition accuracy are greatly improved, the high-efficient structured management of the massive video data is realized, the effective information resources are formed, and the retrieval speed is greatly improved, especially the intelligent real-time detection and recognition of the two mainobjects in the video, namely, the human and the vehicle.

A storage controller, such as a RAID controller arbitrates storage tasks between a hard disk drive and a solid state drive based on predetermined factors, such as the type of information associated with a read or a write or the power available for running the storage devices. For example, a RAID controller on a portable information handling system performs writes and reads for sequential information with a hard disk drive. If power is limited, such as from a battery, the storage controller powers down the hard disk drive and performs storage tasks with the solid state drive with periodic power ups of the hard disk drive to mirror stored information.

The invention discloses a CNN model, a CNN training method and a vein identification method based on CNN. The CNN model comprises multiple convolution layers, a full connection stair layer and a SoftMax layer. In a CNN training process, a database is firstly expanded and multiple biological feature databases including similar features are combined to carry out a training of a model; the full connection layer and the SoftMax layer serve as a multi-classification classifier together; a multi-classification neural network is trained so that the neural network is allowed to learn features capable of identifying vein features; and after the training is finished, the previous layer of the full connection layer is output to serve as feature, and similarity of a pair of images is measured by calculating cosine distance of the features. According to the invention, multi-mode biological feature data is fused and used for training a network, so a problem of insufficient training samples is solved and retrieval speed can be greatly improved in a super large identity identification database.

The invention discloses a remote sensing image retrieval method based on feature selection and semi-supervised learning. In the method, an optimal color feature and an optimal textural feature are selected respectively by utilizing a clustering method according to a minimum description length criterion and an improved Davies-Bouldin index; and then an appropriate semi-supervised learning method is selected according to the binarization weight of the optimal color feature and the optimal textural feature for carrying out remote sensing image retrieval. Compared with the traditional remote sensing image retrieval method, the invention not only can greatly improve the retrieval quality, but also can reduce the calculate quantity in the retrieval process and improve the retrieval speed.

Provided are methods, apparatus and computer programs for enhanced access to resources within a network, including for controlling use of bandwidth-sensitive connections within a network and/or for automated recovery. Hash values are used as ‘unique’ identifiers for resources distributed across a network, and each one of a set of pool servers store the hash values for a set of computers within a LAN. When a resource is required, a hash value representing the resource can be retrieved and compared with hash values stored at a pool server to determine whether the pool server holds a matching hash value. Any such matching hash value found on the pool server represents an identification of a local copy of the required resource, because of the uniqueness property of secure ash values. The information within the pool server can be used to access the required resource. If a large resource such as a BLOB or new version of a computer program can be obtained from another computer within a LAN, a reduction in reliance on bandwidth-sensitive Internet connections and reduced load on remote servers becomes possible.

A structured prediction model learning apparatus, method, program, and recording medium maintain prediction performance with a smaller amount of memory. An auxiliary model is introduced by defining the auxiliary model parameter set θ^(k) with a log-linear model. A set Θ of auxiliary model parameter sets which minimizes the Bregman divergence between the auxiliary model and a reference function indicating the degree of pseudo accuracy is estimated by using unsupervised data. A base-model parameter set λ which minimizes an empirical risk function defined beforehand is estimated by using supervised data and the set Θ of auxiliary model parameter sets.

The invention discloses a spatial data double cache method and a mechanism based on a key value structure, and belongs to the technical field of spatial data storage and management. A double cache mechanism of memory caching and file caching is disclosed by the spatial data double cache method and the mechanism based on the key value structure, the memory caching is first level caching, uses B+tree organizing data, and is written in the file caching by adopting a caching write-back mechanism in an asynchronous mode; the file caching is second level caching, uses large files to be built, and builds caching index based on the B+tree so as to accelerate the speed of searching; and a free space of the file caching uses free space management based on the B+tree to manage. The spatial data double cache method and the mechanism based on the key value structure have the advantages of being free in key value storage mode, fast in searching speed, high in concurrency performance and the like. Storage and visiting efficiency of spatial data caching in network environment are improved, and the spatial data double cache method and the mechanism based on the key value structure can be used for caching of generic spatial data such as remote-sensing images, vector data and dynamic effect model (DEM) in a network geographic information system (GIS).

The invention discloses a method and device for searching multi-media programs. The method comprises the following steps of receiving a request for the multi-media programs sent by a terminal, wherein the types of the multi-media programs include at least one of film and television programs, songs, games and recorded videos, and the request for the multi-media programs comprises program key words at least in one type of Chinese characters, letters and numbers; and searching the multi-media programs matched with the program key words from a data source corresponding to the types of the program key words, and sending the searched multi-media programs matched with the program key words to a display device of the terminal. By means of the method and device for searching the multi-media programs, flexibility and searching efficiency of a system are improved, and user experience is further improved.

The invention discloses a distributed search method and a distributed search system. The method comprises that: a search node receives search conditions input by a user through a client browser, processes the search conditions to generate query tasks, and sends the query tasks to an index control node; the index control node sends the query tasks to index nodes in an index node cluster; the index nodes query index files stored in the nodes according to the received query tasks and return the query results to the index control node; the index control node returns the received query results to the search node; and the search node merges the received query results and sends the merged query result to a client. A distributed structure is adopted in the index node cluster, and the index nodes in the index node cluster can search and query the index files thereof during searching, so parallel search and query are realized, the search speed and the search efficiency are greatly improved, and the search result is timely returned to the user.

The invention provides an address matching method and device. The address matching method includes acquiring a query term inputted by the user; dividing the query term into at least one index keyword; according to corresponding relation of preset index keywords and address information, acquiring address information corresponding to the index keywords one to one; displaying at least one of the address information. The address matching method and device allow higher searching speed.

The invention relates to a voice keyword retrieval method based on an audio template. The method comprises the following steps: to begin with, converting a voice sample template and voice to be retrieved into a probability distribution sequence; then, carrying out matching on the voice sample template and the voice to be retrieved through dynamic time warping to obtain keyword starting and ending time point in the voice to be retrieved and acoustic confidence measure scores of each appearance position; and finally, carrying out warping on the scores obtained by different voice sample templates, and carrying out ranking to obtain a retrieval result. In the retrieval process, information of a specific language is not required at all, thereby maximizing universality and transportability; and meanwhile, operation amount in the retrieval process is reduced, and keyword retrieval speed is accelerated.

The invention relates to a method for realizing the quick retrieval of mass videos. The method comprises the following steps: respectively extracting spatial feature vectors from all frame video images in a video stream of a video library to obtain video feature sequences; extracting key feature vectors from the spatial feature vectors; establishing a distributed storage index database according to the key feature vectors of all video files in the video library; extracting key feature vector sets of videos to be retrieved and extracting video index files of the videos to be retrieved; performing the video similarity retrieving in the distributed storage index database according to the video index files of the videos to be retrieved and outputting video retrieval results of the video files with the similarity larger than the preset value of the system. Through the adoption of the method with the structure, representative visual words are adopted to replace key frames, video information is completely represented, a large amount of redundant of video information does not exist, the video information is very compact, the retrieval speed is increased, and the method has mass data concurrent processing capacity, and is wider in application range.

The invention discloses a method of retrieving the product information of a corresponding food material according to the menu information. According to the method, menu marking information and food material information are input to a database in advance; the two kinds of information are related to each other; the production information is input to the database; the production information is classified according to the food material information; before food material retrieval is carried out, a retrieval request of a user is received firstly, the food material information is read according to the menu marking information, the production information is searched according to the needed food material information, and the production information which is obtained by search is provided for the user to select and purchase. According to the method, the automatic identification process of the menu is automatically completed, the accuracy and the integrity of information are guaranteed, and the trouble that the user needs reading the menu is avoided; the process of inputting the food material retrieval for many times by the user is avoided, the retrieval speed and the process efficiency are greatly improved, terms to be selected of the product information of the food material can be obtained through only inputting the menu marking information by the user in the whole process, and the use is very convenient.

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 provides a method and a system for retrieving a specified object based on multi-feature fusion. The method comprises the following steps of: acquiring an image to be retrieved; carrying out a template matching between the image to be retrieved and a sample image containing the specified object based on color to determine degree of color similarity, and processing the next step if the degree of color similarity exceeds a set threshold; making a comparison on minor features between the image to be retrieved and the sample image and determining degree of similarity of the minor features, wherein the minor features include at least one of texture feature and shape feature; and finally, making a comprehensive judgment and acquiring a comprehensive degree of similarity between a retrieval object in the image to be retrieved and the specified object. The result can also be obtained by extracting minor features from a matching area of the image to be retrieved and making a comparison between the extracted minor features and minor features of the sample image containing the specified object. By a cascaded approach, a similar region between the sample image and the image to be retrieved from coarse to fine can be gradually and accurately determined, the effect of fast, efficient and accurate retrieval is achieved, and manpower and material resources are saved.

The invention relates to a method and a device for increasing a retrieval speed in a database retrieval system and belongs to the technical field of database retrieval. The method comprises the following steps: establishing a data cache in an internal storage; after obtaining an index word, searching in the data cache; if a cache record of a corresponding index word is found in the data cache, generating and returning a retrieval result set according to the cache record and ending the retrieval at this time; if the cache record is not found in the data cache, retrieving the database and returning the retrieval result set; and if the retrieval result set is not empty, writing the index word and the retrieval result set into the data cache. The invention also discloses the device for increasing the retrieval speed in the database retrieval system. In the mode of caching after using, the device reduces the system resource consumed by establishing the cache and increases the use ratio of the cache. In the mode of caching the result set, the caching mechanism is more convenient and quick.

The invention relates to a system and a method for image search through images of multi-task portal vehicles. According to the method, a deep neural network is utilized to establish a multi-task positioning network and a multi-task feature extraction network; the positioning network is trained based on an improved edge box detection technology and cascaded loss functions; positioning and feature detection are performed on vehicles, annual inspection signs and lamps in portal vehicle images, and global features and local features are combined; a softmax loss function and a triad loss function are adopted for network training; and finally local feature vectors are subjected to weighting combination, and global feature vectors of the last full-connection layer of the neural network are utilized to serve as vehicle features for retrieval, wherein an improved k-means algorithm is adopted to find a K class in retrieval, and then an SVM is utilized to form a Hash function for Hamming coding.In this way, retrieval speed is increased, and storage space is saved.