58 results about "Vocabulary tree" patented technology

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.

The invention discloses a privacy-protection index generation method for mass image retrieval, relates to the privacy protection problem in mass image retrieval and involves with taking privacy protection into image retrieval. The method is used for establishing an image index with privacy protection, and therefore, the safety of the privacy information of a user can be protected while the retrieval performance is guaranteed. The method comprises the steps of firstly, extracting and optimizing SIFT (Scale Invariant Feature Transform) and HSV (Hue, Saturation and Value) color histogram, performing feature dimension reduction by use of a use of a manifold dimension reduction method of locality preserving projections, and constructing a vocabulary tree by using the dimension-reduced feature data. The vocabulary tree is used for constructing an inverted index structure; the method is capable of reducing the number of features, increasing the speed of plaintext domain image retrieval and also optimizing the performance of image retrieval. The method is characterized in that privacy protection is added on the basis of a plaintext domain retrieval framework and the inverted index is double encrypted by use of binary random codes and random projections, and therefore, the image index with privacy protection is realized.

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 visual-content-based method for establishing a multi-level semantic map. The visual-content-based method comprises the following steps: gathering images shot by a robot wandering in an environment and labeling the scenes of spots for photography; constructing a hierarchical vocabulary tree; constructing a knowledge topological layer so as to grant knowledge to the knowledge topological layer; constructing a scene topological layer; constructing a spot topological layer. According to the visual-content-based method, a visual sensor is utilized for constructing the multi-level semantic map for a space, and digraph structure is used on the knowledge topological layer for storing and inquiring the knowledge, so that unnecessary operation can be eliminated in a knowledge expression system, and the inserting and inquiring speed is quick; the scene topological layer is utilized for carrying out abstract division on the environment so as to abstractly divide the whole environment into subdomains, so that the image searching space and the path searching space can be reduced; the spot topological layer is utilized for storing specific spot images, the self-positioning can be realized by adopting image searching technology, and the error accumulation problem of self-positioning estimation is solved without maintaining the global world coordinate system.

Described is a technology in which an image retrieval system is updated incrementally as new image data becomes available. Updating is incrementally performed and only triggered when the new image data is large enough or diverse enough relative to the image data currently in use for image retrieval. Incremental updating updates the leaf nodes of a vocabulary tree based upon the new image data. Each leaf node's feature frequency is evaluated against upper and / or lower threshold values, to modify the nodes of the tree based on the feature frequency. Upon completion of the incremental updating, a server that performed the incremental updating is switched to an active state with respect to handling client queries for image retrieval, and another server that was actively handling client queries is switched to an inactive state, awaiting a subsequent incremental updating before switching back to active

A system provides 3D reconstruction using multiple pipelines that run in parallel and each include a worker and a queue of abstract jobs. The system receives a new image in a stream of images by an image processing pipeline, and performs image undistortion on the new image by the image processing pipeline to obtain an undistorted image. The system then detects features in the undistorted image by the image processing pipeline, injects a dictionary search job with the undistorted image and the features into a search pipeline, and injects an image compression job with the undistorted image into a maintenance pipeline. The system then performs multi-image feature matching by the image processing pipeline using a vocabulary tree database to obtain matched images, and injects a structure integration job with the undistorted image and the matched images into a constructive pipeline for camera position and point cloud estimation.

The invention discloses a data-clustering-based adaptive image SIFT (Scale Invariant Feature Transform) feature matching method, which comprises the following steps of: (1) acquiring a reference image sequence, extracting image SIFT feature sets and clustering all image SIFT feature sets by adopting a k-d (k-dimensional) tree; (2) organizing all the image SIFT feature sets by adopting a cascaded vocabulary tree and carrying out secondary feature clustering on the feature sets contained in vocabulary tree nodes; (3) matching BBF (Best Bin First) purified based on a specific value with a dual-mode clustering feature based on comentropy by using the cascaded vocabulary tree to finish the first-stage feature matching; and (4) finishing the second-stage feature matching by using a key image and combining the feature matching results of the two stages; and finally rejecting exterior points by adopting RANSAC (Random Sample Consensus), basic matrix and other technologies. By using the matching method, the robustness of the feature matching can be greatly improved and the adaptability of the feature matching can be effectively enhanced.

A method and system for video-content based retrieval is described. A query video depicting an activity is processed using interest point selection to find locations in the video that are relevant to that activity. A set of spatio-temporal descriptors such as self-similarity and 3-D SIFT are calculated within a local neighborhood of the set of interest points. An indexed video database containing videos similar to the query video is searched using the set of descriptors to obtain a set of candidate videos. The videos in the video database are indexed hierarchically using a vocabulary tree or other hierarchical indexing mechanism.

The invention discloses a target matching method among multiple cameras based on multi-feature fusion and incremental learning. The method comprises the steps that a feature model of a target relates to an SIFT (Scale Invariant Feature Transform) feature of an extracted target; the feature is quantized on an established hierarchical vocabulary tree to form a hierarchical vocabulary tree histogram feature; a color histogram feature is extracted; a preliminary fusion feature is obtained according to the two histogram features; kernel PCA (Principal Component Analysis) dimensionality reduction is conducted on the fusion feature; a nonlinear fusion feature is extracted; classification and identification of the target are that a multi-target nonlinear fusion feature is sent into a multi-class SVM (Support Vector Machine) classifier for the classification and the identification; on-line updating of a target model is accomplished by conducting the incremental learning on the multi-class SVM classifier; and when a new target appears in visual fields of the cameras, and the appearance and the shape of the target are changed greatly, the target model is updated continuously by the incremental SVM learning. The method fuses the vocabulary tree histogram feature of the target with the color histogram feature, and increases an identification rate of the target significantly.

The invention relates to a streetscape retrieval and identification method of an intelligent mobile terminal. The method comprises the following implementation steps: (1) establishing a streetscape image database and a hot-spot information database for different position areas respectively; (2) establishing a vocabulary tree index structure for an image of the database; (3) uploading GPS (Global Positioning System) position information through mobile equipment, and acquiring a hot-spot area around a current position; (4) uploading a streetscape image, performing retrieval by a server-side program according to the uploaded GPS position information and image information, and returning needed contents; (5) displaying the contents on mobile equipment. According to the method, the hot-spot area can be visually displayed on a map, abundant labeling information is acquired, and therefore the user experience is improved.

The invention relates to an unmanned aerial vehicle aerial image matching method based on vocabulary tree blocking and clustering. Firstly, images in a scene are quantified by using a vocabulary tree, and hierarchical clustering is established for the centralized mass features of the images, so that rapid similarity screening of a to-be-matched image set and a mass image set is realized, a rapid scene classifying process is realized, and frame-after-frame matching and selection of the to-be-matched image set according to a traditional method are avoided. Secondly, a thumbnail of two frames of obtained images having similarity is established, and rough matching is performed on the images under the thumbnail. Then, image blocking is performed by using a clustering method, which is an effective trial for the thought of rough-to-fine matching. Furthermore, the unmanned aerial vehicle aerial image matching method is provided for the first time specific to the data characteristics of large data size of unmanned aerial vehicle aerial images, high image resolution, low image overlapping ratio and the like, so that the unmanned aerial vehicle aerial image matching accuracy and efficiency are effectively increased. The effectiveness of the method is verified by testing aerial images in a PAMView: Providence Aerial MultiView Dataset database.

The invention discloses a search processing method and a search processing device. The search processing method comprises the following steps: according to a received retrieval character string, generating a plurality of alternative character strings, wherein preset edit distances are formed between the alternative character strings and the retrieval character string; respectively finding the alternative character strings through a vocabulary tree of an alternative vocabulary; if the alternative character strings are found, supplying the alternative character strings as recommended retrieval character strings to a user. By the search processing method, according to the preset edit distances, a controllable number of alternative character strings are generated, so that the calculation amount of an algorithm is relatively constant and does not increase with increase in the number of the character strings of the alternative vocabulary; in addition, the edit distances between the generated alternative character strings and the character strings in the alternative vocabulary do not need to be calculated one by one, but the vocabulary tree with a relatively high search speed is used for further screening the alternative character strings to obtain the recommended retrieval character strings, so that the search processing speed is increased.

The invention discloses a retrieval method using a random quantization vocabulary tree and an image retrieval method based on the random quantization vocabulary tree. The method comprises the steps that (1) a nearest neighbor search tree is generated, and all feature vectors of a whole database are used as root nodes of a first segment for downward segmentation; (2) at the second level, k points are randomly selected from the whole database to serve as cluster centers, then each feature vector is distributed to the cluster center closest to the feature vector according to a selected similarity measurement method, the whole database is divided into k subsets, and downward segmentation is continued; (3) at the third level, k feature points are randomly selected from a feature vector pool of all k clusters obtained from the second level to serve as cluster centers of the next level; and (4) the steps are repeated. Through the image retrieval method, the problem that in the prior art, vocabulary tree establishment needs a large amount of time is solved, the vocabulary tree can be established in a short time, and the real-time requirement is met.

A location recognition technique that involves using a query image to identify a depicted location is presented. In addition to the query image, there is also a pre-constructed database of features which are associated with images of known locations. The technique matches features derived from the query image to the database features using a specialized vocabulary tree, which is referred to as an informative feature vocabulary tree. The informative feature vocabulary tree is specialized because it was generated using just those database features that have been deemed informative of known locations. The aforementioned matching features are used to identify a known location image that matches the query image. The location associated with that known location image is then deemed to be the location depicted in the query image.

A method for multi-subtree-based distributed image training and searching comprises the following steps: step 1, selecting initial cluster central points and clustering by computational nodes, and distributing the clustered new cluster central points to each computational nodes; step 2, training task subtrees by each computational node, and taking a cluster central point as a task subtree growing point. The method for multi-subtree-based distributed image training and searching further comprises step 3, extracting feature points from the image to be searched, and transmitting the feature points to corresponding computational nodes according to cluster central point affiliation; step 4, utilizing task subtree to process and transmitting the results to management nodes by the computational nodes, and summarizing the computational results of computational nodes to acquire image searching results by the management nodes. The method can divide the training task of one vocabulary tree into the training tasks of a plurality of subtrees, enable a plurality of computational nodes to process in parallel so as to accommodate larger image training set for strong expansibility, meanwhile reduce the time cost for process of image training and searching.

The invention discloses an image retrieval method based on incremental learning of a large vocabulary tree. The generated large training picture vocabulary tree which contains millions of images is used as a dictionary, leaf node number information associated with feature descriptors of newly added pictures is extracted from the vocabulary tree so as to form a vector for description, the incremental learning of the newly added pictures is completed, leaf node number information associated with feature descriptors of to-be-queried pictures is extracted in a same mode as the incremental learning so as to form a vector, and the two leaf node information vectors are compared so as to find out an image represented by the vector with a highest contact ratio with the leaf node number information of the to-be-queried pictures as a retrieval result to be returned. The image retrieval method has the advantages of high accuracy of the retrieval result and good robustness, and a real-time image retrieval task of a database increment can be satisfied.

The invention provides a network behavior detecting method and device and belongs to the field of network auditing. The method and device is characterized in that a plurality of buffer areas are arranged through the collecting buffer area transverse expansion technology, high-speed data collecting under a 10 gigabit network environment is achieved, a network communication behavior feature knowledge database is built on the basis of a tree-like logic model, a finite automata is built and determined according to four vocabulary trees, respectively corresponding to a data link layer, a network layer, a transmission layer and an application layer, obtained after the analysis of the network communication behavior feature knowledge database so as to improve a multi-mode matching algorithm AC-BM, and fast matching of network communication behavior features is achieved. Compared with traditional data packet detecting and data auditing technologies, the network behavior detecting method and device is high in feature matching efficiency, low in algorithm complexity, reasonable in knowledge database design and beneficial to upgrading and expansion.

The invention discloses a GPU acceleration-based large-scale image vocabulary tree retrieval method and belongs to the field of computer vision. Based on structural features of a GPU platform, a novel vocabulary tree data structure is employed; and traditional SIFT feature points are modularized and organized, so GPU large-scale parallel computing capacity can be fully utilized and large-scale image SIFT feature point mapping efficiency can be greatly improved. After batches of SIFT feature points are mapped into a histogram, a quick and compact method is employed to enhance and improve an original reverse index scheme, so post vision vocabulary peculiarity express, histogram normalization and query image grades can be accelerated to a certain extent. The invention further provides a GPU acceleration-based large-scale image vocabulary tree retrieval system, playing an important acceleration role for image retrieval and having a great effect on a large-scale scene three-dimensional reconstruction algorithm efficiency increase.

The invention discloses an unmarked augmented reality multi-target registration tracking method. The method comprises the following stages: 1) an offline stage of performing hierarchical k-means clustering to obtain a vocabulary tree model, registering corresponding image ids for reverse indexes of all leaf nodes of a vocabulary tree, and finally updating the vocabulary tree to a vocabulary tree with tf-idf according to occurrence frequencies of the leaf nodes of the whole tree and a total to-be-registered image quantity; and 2) an online stage which is a real-time response system and comprises the steps of retrieving most similar images according to a trained vocabulary tree for images input in real time by a camera, calculating out to-be-loaded 3D object initial poses through pose estimation of the camera, tracking motions of targets by adopting a KLT tracking algorithm and finally creating augmented reality scenes of the targets through a rendering thread. The method provides an efficient and reliable solution for unmarked multi-target augmented reality.

The invention discloses a cloud image recognition method based on vocabulary tree retrieval and similarity verification. The method includes the steps of image obtaining, image uploading, image recognizing and similarity verification, wherein in the image obtaining step, a target image is obtained, all ORB feature points of the target image are extracted through the ORB algorithm to generate an ORB description subsequence of the target image; in the image uploading step, the ORB description subsequence is uploaded to a cloud image database based on description subsamples; in the image recognizing step, the cloud image database conducts matching recognition on the image through the retrieval algorithm based on a vocabulary tree, and N candidate images with higher matching degree are fed back; in the similarity verification step, the candidate images are found in the cloud image database to obtain 128 dimensional vectors of the target image and the candidate images, the distances between the target image and the candidate images are calculated respectively, and the candidate image with the shortest distance is found out. By means of the method, influences of a poor network on recognition speed are small, the retrieval speed is high, and the retrieval precision is high.

Disclosed is a method for constructing a lexical tree for speech recognition, wherein, even though a name included in an address book in a communication device such as a cellular phone and a word such as “house / office / cellular phone” are sequentially and successively uttered, the method allows the uttered speech to be precisely recognized. The method for constructing a lexical tree constructs a lexical tree including a name tree composed of names included in an address book in a communication device and an expansion vocabulary tree composed of words following the names, respectively.

The invention discloses a digital image processing method applied to an intelligent early education server. The method includes the steps of receiving image data sent by a client side, extracting feature information of the image data, matching the feature information with tree species features in a preset vocabulary tree, obtaining a prebound multimedia file corresponding to the tree species features when matching succeeds, and sending the multimedia file to the client side to be played on the client side. The invention further discloses another digital image processing method, the intelligent early education server and the intelligent early education client side. Interactivity and interestingness of early education can be improved.

The invention discloses a method for image identification based on vocabulary tree retrieval and brute-force matching. The method comprises an image obtaining step of obtaining an image, using an ORB algorithm on the image to extract all ORB feature points, and generating a corresponding descriptor for each ORB feature point, and generating an ORB description subsequence of the image; an imaging uploading step of uploading the ORB description subsequence into a cloud image database; an image identification step: in the cloud image database, utilizing a search algorithm of a search vocabulary tree to perform matching identification on the image and returning N candidate images with matching scores among the top few; a brute-force matching step of finding the candidate images in the cloud image database, and performing one-to-one brute-force matching on the candidate images and the ORB description subsequence of the image to determine an optimal matching image. Search matching is implemented through extraction of the image descriptor, the influence on an identification speed caused by a poor network is small, and searching precision is high under the conditions that the size of the vocabulary tree structure is limited.

The invention discloses an unmanned aerial vehicle image matching pair selection method and system based on vocabulary tree retrieval, and the method comprises the steps: firstly carrying out SIFT feature point extraction of each image in an image subset obtained through sampling through employing a random sampling strategy and an SIFT algorithm; secondly, for extracted SIFT feature points, constructing an initial feature set, sorting all SIFT feature points in the initial feature set from large to small according to scales, selecting the first h feature points in the sorted set, and constructing a feature subset about the h feature points; thirdly, for the obtained feature subset, constructing a vocabulary tree for unmanned aerial vehicle image retrieval by adopting a hierarchical K-meansclustering algorithm; wherein the vocabulary tree is used for establishing an unmanned aerial vehicle image index; and finally, carrying out unmanned aerial vehicle image matching pair selection through an adaptive similar image quantity selection algorithm by utilizing the established unmanned aerial vehicle image index.

The invention provides an image retrieval method based on vocabulary tree information fusion and Hausdorff distance combination and belongs to the technical field of image retrieval. The image retrieval method comprises the steps that firstly, images to be retrieved and SIFT characteristics of an image library are extracted, an SIFT descriptor histogram and SIFT descriptor kernel density are generated, then the SIFT descriptor kernel density and the SIFT descriptor histogram are fused, and finally an improved Hausdorff distance is used for image matching by improving traditional Hausdorff distance measure. The image retrieval method includes information fusion of extensible vocabulary trees based on SIFT kernel density and extensible vocabulary trees based on the SIFT histogram, the improved Hausdorff distance in image similarity measurement standard and combination of information fusion and the improved Hausdorff distance. Experiments show that the method can not only improve the image retrieval accuracy and is applicable to retrieval of images with complex backgrounds.

The invention discloses a three-dimensional object retrieval method based on multi-view classification. The method comprises the steps of obtaining an original image; processing the original image toobtain a multi-view image; processing the multi-view image to obtain a feature descriptor; processing the feature descriptor to obtain a vocabulary tree; and retrieving a to-be-retrieved image in thevocabulary tree to obtain the serial number of the original image. According to the method, a multi-view classification algorithm is adopted to obtain an omnibearing multi-view feature image of the three-dimensional object in a two-dimensional plane, the three-dimensional object can be directly retrieved, the method is simple, convenient and easy to implement, and the problem that the retrieval performance is reduced due to view angle changes of the object shot by a user in three-dimensional image retrieval is solved.

The invention discloses a dynamic self-adaptive binary hierarchical vocabulary tree image retrieval method, comprising the following steps: (1) extracting features; (2) constructing a binary hierarchical balance vocabulary tree; (3) selecting two centroids in the space as segmentation points each time, and segmenting the whole space by a hyperplane perpendicular to a straight line passing throughthe two centroids until all the points are segmented; (4) describing each subspace through the centroid of the subspace; dispersing the extracted N descriptors on a node of a binary tree of which thedepth is log2N + 1; (5) performing feature extraction on a picture to be retrieved to obtain a descriptor set P * M which represents P feature vectors; (6) sequentially traversing the binary hierarchybalance vocabulary tree by the P feature vectors and retrieving; and (7) according to the inverted index, each node having a picture ID corresponding to the node, and taking intersection of picture ID sets corresponding to leaf nodes obtained by P feature vectors through retrieval traversal to obtain a picture ID to be retrieved.

The invention discloses an image retrieval method based on a CNN (Convolutional Neural Network) feature vocabulary tree, and aims to solve the technical problem of low accuracy in an existing vocabulary tree method. The method comprises the following implementation steps that: firstly, generating the derivative image of each image in an image library, and extracting the CNN feature of each image in the image library; according to the extracted CNN feature, constructing the CNN feature vocabulary tree; then, generating the derivative image of each image to be retrieved, and extracting the CNN feature of each image to be retrieved; comparing the path of the CNN feature of each image to be retrieved with the path of the CNN feature of the relevant image of the image to be retrieved in the CNN feature vocabulary tree; calculating a distance between the image to be retrieved and the relevant image, and combining the distance between the image to be retrieved and the relevant image with initial similarity; and finally, according to the comprehensive similarity of each image to be retrieved and the relevant image, outputting the retrieval result of each image to be retrieved. The method is high in image retrieval accuracy and can be used for a medical image computer-assisted diagnostic system and a search-by-image system.