254 results about "Object label" patented technology

The invention provides a video content quick retrieving method based on an object tag. The method comprises the following steps: extracting and analyzing the color feature, contour feature, scene feature and character feature of a moving object in each image frame of a video; processing a plurality of pictures of known types by using the feature extraction method, and training a contour classifier and a scene classifier by using the contour features and scene features of the pictures; processing a video to be retrieved by using the feature extraction and analysis method and the classifiers soas to generate type tags of objects in each image frame of the video, wherein the type tags are used for constructing an object tag database; and retrieving a response server to search the object tagdatabase to find videos related to a query request submitted by a user, and generating an ordered result for the user to browse and refer. The method provided by the invention can be used for retrieving the video content at a speed similar to that of the conventional text retrieval only by searching the object tag database and achieving the fine granularity retrieval of the video content, so thatthe method is more accurate than the conventional method.

An ultra wideband (UWB) or short-pulse RF system is disclosed that can be used to precisely locate or track objects (such as personnel, equipment, assets, etc.) in real-time in an arbitrarily large, physically connected or disconnected, multipath and / or noisy environment. A system implementation includes multiple zones or groups of receivers that receives RF signals transmitted by one or more timing reference tags and one or more objects having associated object tags. Each zone or group may share a common receiver. By combining a multiple reference tag system with a virtual group of receivers, i.e., a zoning technique or system, a cost-effective system can be provided that offers scalability and flexibility to monitor a significantly expanded coverage area.

A system of tracking objects through a service process at a service shop comprises a computer network configured to track movement of tagged service objects from physical location to physical location within the service shop. The computer network comprises tag readers with antennae configured to receive transmissions from tags encoded with identifiers and connected to associated tagged service objects. The tag readers transmit identifiers obtained from the tag transmissions together with an indication as to which antenna received each transmission. Servers receive the identifiers transmitted from the tag readers and determine, based at least on the identifiers and which tag reader antenna received the transmissions, which physical location each service object associated with the received identifiers occupies. Client workstations generate a visual map of physical locations together with summary status information showing which service objects occupy the displayed physical locations.

An object control and tracking system and related methods using separate object identification and location detection mechanisms for objects, such as keys, that are maintained in a secure enclosure, such as a key drawer. A plurality of object slots are located on the top tray of the enclosure to receive object tags that include both an RFID tag and an object to be tracked. The objects (e.g., keys) are attached to a portion of the object tags that are outside of the enclosure. Presence detectors are used to determine if an object tag is present in the corresponding slot of the enclosure. RFID sensors located on opposite interior side walls of the enclosure interrogate each RFID tag to determine the presence of each RFID tag within the enclosure. The object control and tracking system includes a controller, a memory for storing an object control database, and related processing logic operating on the controller to control scanning of the presence detectors to determine an object tag's presence in a corresponding slot, to identify each object present in the enclosure and to compare the identified objects with an object control database to determine each object removed or replaced since the previous database update.

A wireless radio frequency positioning method adopting region division algorism comprises the steps of: arranging reader at the boundary of an object region, distributing the reference labels as a rectangular grid type in the indoor object region; arranging an object with a radio frequency label in the object region, using each reader and their antenna to check the electromagnetic wave of each radio frequency label, transmitting the measured reference label and the radio frequency signal intensity value (RSSI value) of the object label to a host computer; at the host computer, comparing the RSSI absolute difference of each reference label and object label to find four nearest reference labels near the object, preliminarily determining the effective positioning region of the object; using the region division algorism to reduce the effective positioning region formed by the nearest reference labels to 1 / 8; using the residual weighting algorism between different reference labels to attain the position of the object. The method has the advantages of low power consumption and high positioning accuracy.

A wireless radio frequency positioning method based on virtual reference label algorism comprises the steps of: arranging reader at the boundary of an object region, distributing the reference labels as a rectangular grid type in the indoor object region; arranging an object with a radio frequency label in the object region, using each reader to reach the radio frequency signal intensity of each reference label and object label to be transmitted to a host computer to be processed to attain the coordinates of the object. The invention is characterized in that the host computer builds a virtual reference label in calculation, uses the virtual reference label as substrate to induce the concept of similar map; the host computer preliminarily finds the positioning region of the object on the similar map corresponding to each reader; the similar maps are intersected to reduce the positioning region of the object; the residual weighting algorism between the possible positions and the position information of the virtual reference label are used to attain the position coordinate of the object. The method has the advantages of low power consumption and high positioning accuracy.

A process of encrypting an object having an associated object tag includes generating a cryptographic key by binding an organization split, a maintenance split, a random split, and at least one label split (710). A cryptographic algorithm is initialized with the cryptographic key, and the object is encrypted using the cryptographic algorithm (712) according to the object tag, to form an encrypted object. Combiner data is added to the encrypted object (711). The combiner data includes reference data, name data, a maintenance split or a maintenance level, and the random split (710). Alternatively, key splits are bound to generate a cryptographic key, and a cryptographic algorithm is initialized with the cryptographic key. The initialized cryptographic algorithm is applied to the object according to a cryptographic scheme determined by the object tag, to form an encrypted object. One of the key splits corresponds to a biometric measurement.

Machine learning method that learns to convert 2D video to 3D video from a set of training examples. Uses machine learning to perform any or all of the 2D to 3D conversion steps of identifying and locating objects, masking objects, modeling object depth, generating stereoscopic image pairs, and filling gaps created by pixel displacement for depth effects. Training examples comprise inputs and outputs for the conversion steps. The machine learning system generates transformation functions that generate the outputs from the inputs; these functions may then be used on new 2D videos to automate or semi-automate the conversion process. Operator input may be used to augment the results of the machine learning system. Illustrative representations for conversion data in the training examples include object tags to identify objects and locate their features, Bézier curves to mask object regions, and point clouds or geometric shapes to model object depth.

The invention discloses a method for carrying out condition query on a database table. The method comprises the following steps: acquiring the correlation relation between a page and a target database table and analyzing a query object label compiled in a page file to display a query page, wherein the query object is formed by encapsulating a query input frame in the page; extracting a query condition value input in the query input frame by a user from the query object when receiving a query request sent by the user on the query page; binding the extracted query condition value to the target database table correlated to the query page to form a query sentence; and executing the query sentence and returning a query result in accordance with query conditions. The method improves developmentefficiency and quickens the response speed of an application system when a query type changes.

Embodiments of the invention provide a network training method and device, an image processing method and device, a storage medium and electronic equipment. The neural network training method comprises the following steps of: generating a neural network through parameters, obtaining an effect transformation parameter of a first sample image, and transforming the first sample image into a second sample image; obtaining effect classification and detection data of the second sample image through a classification neural network; and training the parameters to generate the neural network accordingto the effect classification and detection data of the second sample image and image effect classification labeling information of the first sample image. On the basis of a generative adversarial network, simple and object labeling data of image effect classification is taken as supervision information of training, accurate image effect parameter data labeling does not need to be carried out on selected sample images, and parameters for generating image effect transformation parameters are trained through a weak supervised learning manner so as to generate neural networks.

A positioning method which is based on the wireless radio frequency identification technology comprises the steps that readers are arranged at the boundary of a zone to be monitored; reference tags are fixed at an indoor zone to be monitored in a rectangular grid type; when an object to be monitored with radio-frequency tags is arranged in the zone to be monitored, each of the readers detects the electromagnetic wave of each radio-frequency tag through the antenna of each of the readers, the radio-frequency signal intensity values of the reference tags which are obtained by the measurement and the tags of the object to be tested are sent to an upper computer; the upper computer compares the absolute difference values of the radio-frequency signal intensity of each of the reference tags and the tags of the object to be tested to determine a plurality of nearest reference tags of the object to be tested; each nearest reference tag and the calculated position of the object to be tested are obtained through the trilateration method, and the average measurement error of a system is obtained by comparing the calculated positions and the actual positions of the nearest reference tags; the calculated positions of the object to be tested adds the average measurement error to obtain the position coordinates of the object to be tested. By utilizing the method of the invention, a positioning system has good accuracy and real-time in the aspect of position induction.

A method for editing Web-based documents and a software package for implementing the method is provided. The method can include receiving from a user an indication of a selected portion of a Web-based document to be edited and of a desired editing function to be performed on the selected portion and inserting immediately prior to the selected portion a first editing tag corresponding to the desired editing function. Also, the method may include detecting object tag elements within the selected portion, inserting immediately prior to each object tag element within the selected portion a second editing tag corresponding to the desired editing function and inserting the second tag at the end of the selected portion, and inserting immediately after each object tag element within the selected portion the first editing tag, wherein the first and second editing tags are distinguishable from the object tag elements.

The invention discloses a wireless radio frequency locating method based on the Bayesian theory. The method comprises the following steps: readers are positioned at the boundary of an area to be monitored; a reference tag is fixed in the indoor area to be monitored in a rectangular grid form; when a radio frequency tag is arranged in an object to be monitored, and the object is positioned in the area to be monitored, the readers transfer the radio frequency signal intensity of the reference tag and the object tag to be monitored to a host computer; the host computer compares the absolute differences of the radio frequency signal intensity of the reference tag and the tag object to be monitored for determining the recent reference tag near the object to be monitored and obtaining the reference position of the object to be monitored; then, the host computer obtains the distance information between the tag object to be monitored and each reader; and finally, the host computer carries out data fusion to the obtained positional information based on the Bayesian theory, thereby obtaining the position coordinates of the object tag to be monitored. By dint of the method, a positioning system has good flexibility and adaptability in restraining the noise interference, and has good precision and real-time performance in the position inductive aspect.

The invention discloses an RFID (Radio Frequency Identification) mutual authentication method based on Hash. The RFID mutual authentication method is used for solving the technical problem of low efficiency of retrieving an object label of a database by using a traditional RFID mutual authentication method. The invention adopts the technical scheme that a pseudo-random number generator, a Hash function and a simple bitwise exclusive-OR operation are adopted; when a server retrieves the object labels in the database, comparison is carried out item by item firstly by using label response message information and label records in the database; and after a matched record is found, calculating validation is carried out. According to the method, calculating validation for each label record in the database is avoided, thereby the calculated quantity of the database of the server is effectively reduced and the efficiency of retrieving the object label in the database is remarkably increased; and in addition, the calculated quantity of the label in the authentication process is also reduced.

Described is a system for multiple-object recognition in visual images. The system is configured to receive an input test image comprising at least one object. Keypoints representing the object are extracted using a local feature algorithm. The keypoints from the input test image are matched with keypoints from at least one training image stored in a training database, resulting in a set of matching keypoints. A clustering algorithm is applied to the set of matching keypoints to detect inliers among the set of matching keypoints. The inliers and neighboring keypoints in a vicinity of the inliers are removed from the input test image. An object label and an object boundary for the object are generated, and the object in the input test image is identified and segmented. Also described is a method and computer program product for multiple-object recognition in visual images.

System for managing a collection of objects, for example in a store, employing identification tags (RFID) assigned to object (O) locations and individual labels (ID-O) associated with the objects. The objects are identified using a portable label scanner (RFIDL) operating in either a local mode (M1) or in a single mode (M2), with said scanner being connected to a control unit (CPU) that controls the operation of the scanner. The local mode (M1) consists of reading the label (ID-Pj) of the display location (Pj) of a batch of objects (Oi) and of scanning the label of the objects located in this location for comparison and verification. The single mode (M2) is a close-up scan of the label of an object, taken individually.

Described is a system for improving object recognition. Object detection results and classification results for a sequence of image frames are received as input. Each object detection result is represented by a detection box and each classification result is represented by an object label corresponding to the object detection result. A pseudo-tracklet is formed by linking object detection results representing the same object in consecutive image frames. The system determines whether there are any inconsistent object labels or missing object detection results in the pseudo-tracklet. Finally, the object detection results and the classification results are improved by correcting any inconsistent object labels and missing object detection results.

An image processing apparatus includes an acquiring unit configured to acquire multiple tracking results about an object tracked in multiple video images captured by multiple imaging units. The tracking results correspond one-to-one to the video images. Each of the tracking results contains a position of the object detected from an image frame of the corresponding video image and a tracking label that identifies the object in the video image. The apparatus further includes a relating unit configured to relate objects, detected from image frames of the video images, based on the tracking results acquired by the acquiring unit to obtain relations and an object label generating unit configured to generate an object label based on the relations obtained by the relating unit. The object label uniquely identifies the object across the video images.

The invention relates to a method for analyzing cells that are present as closed clusters. According to said method, a planar tissue preparation is subjected to an identification staining of the cell nuclei and a target structure staining of cell objects that is different from the identification staining. Digital images are recorded of the stained tissue preparation by means of an electronic image recording device and at least one image of a subsection of the tissue cut is displayed in at least one coloration. According to the inventive method, at least one parameter of the cell nuclei and at least one parameter of the cell objects labeled by target structure staining is restricted to a predetermined range of values. Cell nuclei and cell objects whose parameters correspond to the respective parameter range(s) are detected and optionally displayed using image processing algorithms in the image of said subsection. The image content of at least one image detected for the cell nuclei is correlated with the image content of at least one image detected for the target-structure stained cell objects to detect the individual cells. On the basis of the cell nuclei identified a cell growth or a cell enlargement is induced using a predetermined arithmetic algorithm to reconstruct the individual cells. In doing so it is made sure that neighboring cells do not fuse. The number of reconstructed individual cells is determined and / or the individual cells are divided into populations according to certain parameters.

An apparatus and method for generating object-labeled images based on query images in a video sequence are provided. A video sequence is divided into a plurality of shots, each of which consists of a set of frames having a similar scene, and an initial object region is extracted from each of the shots by determining whether an object image exists in key frames of the shots. Based on the initial object region extracted from each of the key frames, object regions are tracked in all frames of the shots. Then, the object regions are labeled to generate object-labeled images. Therefore, the object-labeled image generating apparatus and method can be applied regardless of the degree of motion of an object and time required to extract query objects is reduced.

A radio frequency positioning method comprises the steps as follows: readers are arranged at the border of a region to be monitored; reference tags are fixed in the indoor region to be monitored in the form of rectangular grids; when an object which is provided with radio frequency tags and ready to be monitored is arranged in the region to be monitored, all the readers detect the electromagnetic waves of all the radio frequency tags through their own antennas, and the radio frequency signal intensity values of the reference tags which are obtained by the measurement and the tags of the object to be monitored are sent to an upper computer; the upper computer compares the absolute difference value of the radio frequency signal intensity of all the reference tags and the tags of the object to be monitored to determine a plurality of nearest reference tags which are in the vicinity of the object to be monitored; and the position of the object to the monitored is calculated according to the position information and the weighting factors of a plurality of nearest reference tags which are in the vicinity of the object to be monitored. A positioning system can have good flexibility and dynamic property, as well as good accuracy and real-time property on the position sensing aspect with the help of the radio frequency positioning method.

The invention discloses a method and a device for positioning a radio frequency identification (RFID) label. The method comprises the following steps of: respectively reading a plurality of linearly-arranged object label presetting times under a plurality of appointed transmitting power by an antenna of a RFID reader; recoding the times that each object label is successfully read by the antenna under each appointed transmitting power; calculating the sum of the times that each object label is successfully read by the antenna; ensuring an arrangement sequence of each object label relative to the antenna by sequencing the sum of the times that each object label is successfully read; calculating normalization reading probability of each object label relative to the antenna; and ensuring the distance of each object label to the antenna by inquiring and reading a probability map according to the normalization reading probability of each object label relative to the antenna, and reading a corresponding relationship between the normalization reading probability of a probability map recording label relative to the antenna and the distance of the label relative to the antenna.

Machine learning method that learns to convert 2D video to 3D video from a set of training examples. Uses machine learning to perform any or all of the 2D to 3D conversion steps of identifying and locating objects, masking objects, modeling object depth, generating stereoscopic image pairs, and filling gaps created by pixel displacement for depth effects. Training examples comprise inputs and outputs for the conversion steps. The machine learning system generates transformation functions that generate the outputs from the inputs; these functions may then be used on new 2D videos to automate or semi-automate the conversion process. Operator input may be used to augment the results of the machine learning system. Illustrative representations for conversion data in the training examples include object tags to identify objects and locate their features, Bézier curves to mask object regions, and point clouds or geometric shapes to model object depth.

The embodiment of the invention discloses a method and device for determining object tags, a method and device for establishing a tag index and a method and device for searching for objects. The method for determining the object tags includes the steps of obtaining first text information corresponding to the first object, and obtaining a first participle word set corresponding to the first objectaccording to the first text information; based on a preset corresponding relationship between the tags and keywords, determining a participle word, successfully matched with the corresponding keywordin the preset corresponding relationship, in the first participle word set and the candidate tags corresponding to the successfully-matched participle word; setting a part of or all the candidate tagsas the target tags of the first object. The efficiency and accuracy of determining the object tags can be improved.

The embodiment of the invention discloses a data processing method and device and a storage medium, wherein the method comprises the steps of separately obtaining the sample object data from multiplepieces of multimedia data, and extracting the object feature information of the sample object data; clustering the plurality of sample object data according to the object feature information to obtaina plurality of first object feature clusters with different cluster labels; cleaning the heterogeneous object data in each first object feature cluster, and determining the cleaned first object feature cluster as a second object feature cluster; performing similar cluster merging among the plurality of second object feature clusters, generating a plurality of third object feature clusters with different cluster labels, and updating the object labels of the sample object data into cluster labels of the third object feature clusters to which the object labels belong; and based on the object label and the object feature information of the sample object data in each third object feature cluster, training an object recognition model. According to the invention, the accuracy of the face recognition can be improved.

The invention provides an aligning system and method based on image technique. And the system is arranged in a projection exposure device composed at least of light source, mask, mask bearing table, mask bearing table motion controller, optical projection system, exposed object, plate bearing table and plate bearing table motion controller, and comprises at least two position aligners uniformly distributed by using the optical axis of the optical projection system as axis, mask label arranged on the mask, exposed object label arranged on the exposed object, and reference label arranged on the plate bearing table, where the position aligners detect and transmit position information of various labels to the motion controllers to control position adjustment of the mask, exposed object and plate bearing table to align them, and the aligning system is equipped with an aberration compensation lens and a lens calibrating unit corresponding to each position aligner. And the invention can reduce system manufacturing cost and system complexity by adopting narrow waveband design mode and cooperating with the aberration compensation lens.

The invention provides an application traffic analyzing method and belongs to the technical field of computers. The application traffic analyzing method comprises: acquiring access logs of an application to be analyzed; based on a pre-set funnel configuration file of the application, matching each access log, determining a log index table according with the access logs of the funnel configuration file; based on a path list obtained by decomposing an access path to be analyzed, matching an actual access path determined according to the log index table; calculating a traffic analyzing index of the application to be analyzed according to a matching result. By adopting the application traffic analyzing method, the problems in the prior art that a presentation rate or a click rate is calculated based on frequent path excavation and macroscopic analysis cannot be carried out on pages or modules with the same type are solved; the same analysis object labels are set through the pages with the same type, and data of the pages with the same type can be clustered and analyzed to obtain a macroscopic analysis result.