1370 results about "Target type" patented technology

The present invention is in general related to target type estimation in target tracking systems. The invention enables a low complexity estimation of target type, using, e.g. ESM sensor data, by the introduction of an ambiguity restoring procedure in certain likelihood calculations. The invention further enables the systematic use of target type probability information in the calculation of strobe track crosses and their associated quality which is particularly useful for deghosting purposes. Methods for utilisation of target type probability information in the processes of strobe tracking, association, track quality evaluation and multiple hypothesis tracking are also disclosed.

In a firmware update apparatus, a reference device specification part specifies one of a plurality of audio processing devices as a reference device. A version number acquisition part acquires version numbers of firmwares stored in the audio processing devices. A matching table storage stores a matching table defining a combination of compatible version numbers of firmwares from among combinations of version numbers of firmwares. An incompatible device display part performs display for specifying an audio processing device that stores a firmware that does not satisfy compatibility defined in the matching table with a firmware stored in the reference device. An update target type selection part specifies a firmware update target device type from among a plurality of device types. A firmware update instruction part transmits an update command instructing update of the firmware to one or more of audio processing devices belonging to the firmware update target device type selected by the update target type selection part.

A method and apparatus are disclosed that enable an over-the-horizon-radar (OTHR) system to detect and track multiple target classes simultaneously, where target classes are defined by the speed and acceleration of the tracked objects. The OTHR is tasked in a staring mode, with a bandwidth and waveform repetition frequency that enable detection of Doppler shifts from all target types, with sufficient clutter reduction and range resolution. The backscattered echoes are buffered for each target class and processed independently. The output of an automatic tracking algorithm then preferentially plots target progress on a single digital map for all target classes.

The present disclosure relates to various apparatus, systems and methods of identifying and treating tissue using at least one electrical property of tissue. Provided is a method for identifying and treating tissue, the method including providing a electrosurgical treatment device including an electrode assembly for measuring one or more electrical properties of a target tissue, the electrode assembly being mounted on a distal end thereof, measuring the one or more electrical characteristics of the target tissue, comparing the measured electrical property values of the target tissue against electrical property values of known tissue types, identifying a tissue type of the target tissue, adjusting an energy delivery configuration of the electrosurgical treatment device to the type of target tissue, and activating the electrosurgical treatment device to treat the target tissue.

The invention discloses a federated learning-based recommendation model training method, a terminal and a storage medium. The method comprises the steps of obtaining a user historical behavior data set recorded by a client application in multiple preset types of application projects; extracting a single characteristic user vector of each client application based on each group of user historical behavior data set; extracting a project feature vector set and a project score set from a user historical behavior data set of the target application; combining each single feature user vector, the project feature vector set and the project score set to obtain a local training sample set; and participating in federated learning based on the local training sample set to obtain a recommendation modelof the target type application project. According to the method, model training is carried out under a federal framework to protect user privacy data, meanwhile, recommendation model training is carried out on the basis of multi-scene data, the recommendation model obtained through training can more accurately locate the preference characteristics of the user, and therefore the recommendation effect of the recommendation model is improved.

The invention discloses a target detection and recognition method for water surface panoramic images, and belongs to the technical field of computer visual recognition. The method comprises the following steps of: firstly carrying out target detection and recognition on a panoramic image by utilizing a target detection and recognition neural network model, and obtaining a target type and an initial position of a detection box; carrying image segmentation on a target local area to obtain a plurality of superpixel blocks, and combining the superpixel blocks by adoption of a multi-threshold fusion strategy so as to obtain an adjusted target detection box; calculating an edge map of a new target local area, adjusting the target detection box according to peripheral edge points of the target detection box so as to obtain a final target detection box; and finally converting the position of the final target detection box into a target practical position. According to the method, distortion effect in panoramic images can be effectively overcome, and target positions can be correctly recognized from the panoramic images.

The invention relates to a video detection technology and discloses an intelligent marking, searching and replaying method and device for surveillance videos of shared lanes. According to the invention, when a moving target is detected, the judgment on the type of the target is triggered, and corresponding information is detected according to the target type and is written to a position corresponding to a trigger frame in a video bitstream, therefore, the search of semantic features can be performed conveniently on targets in various types on the shared lanes, and corresponding video frames are positioned fast according to the search result. By virtue of the method and the device, the problem that the videos and the target characteristic information are inconvenient to search and store when being separated from each other is solved; and the trigger begins once a predetermined condition is met, and the video is recorded synchronously, thus, the miss trigger of moving targets can be be avoided, and a storage space required to store captured pictures can be saved as well.

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.

Sensors determine the kinematic measurements of a boosting missile, and the information is applied to a plurality of pairs or sets of filters, one of which is matched to the characteristics of a particular target type, and the other of which is general, and not matched to a particular target, for producing from each filter of the set missile position, velocity, acceleration, and specific mass flow rate states, and covariances of those states. From the filtered information, the estimates are made of at least missile mass flow rate, thrust, velocity at stage burnout, and remaining burn time. The likelihood is computed that the states and covariances from the filter sets represent the same target. The largest likelihood is selected as representing the target. In one mode, the estimated parameters are weighted and summed for a composite likelihood.

The embodiment of the invention provides a split-screen display method and a terminal. The method comprises: detecting starting operation which is triggered aiming at a split-screen mode of the terminal; starting the split-screen mode of the terminal according to the starting operation; under the split-screen mode, setting a display screen of the terminal into at least two split-screen regions according to a pre-set manner; acquiring types of applications to be displayed; acquiring a target application with the type corresponding to a target type from the applications to be displayed; displaying the target application on the pre-set split-screen region; displaying the other applications except the target application in the applications to be displayed on the residual split-screen regions respectively. By implementing the embodiment of the invention, a plurality of split-screen windows can be operated at the same time, the flexibility of screen splitting is improved and user experience is improved.

The invention discloses a data synchronization method and device. The method comprises the steps of detecting synchronization operation, wherein the synchronization operation is used for making a request for synchronizing a target type of data of a first node to a second node; obtaining a synchronization task triggered by the synchronization operation, wherein the synchronization task is used for synchronizing the target type of data stored on slave devices of the first node to the second node, and the first node comprises a master device and a plurality of slave devices on which the data on the master device is backed up; carrying out a synchronization task, reading a target message containing the target type of data from a message queue, wherein the messages cached in the message queue carry various types of data obtained from the salve devices of the first node; and storing the target type of data in the target message to the second node. According to the method and the device, the technical problem that in related technologies, the efficiency is relatively low when data synchronization processing is carried out is solved.

The invention relates to a sea surface micro-motion target Radon-linear contact transformation (RLCT) long-time phase-coherent accumulation detecting method which belongs to the technical field of radar signal processing and detection. The method comprises the following steps: (1) performing demodulation and pulse compression on radar echo, and completing calculation in pulse; (2) pre-judging sea conditions and detection target types and determining long-time phase-coherent accumulation parameters; (3) adopting RLCT compensation distance and Doppler migration and calculating micro-motion target signal energy; (4) viewing and searching parameters and establishing a distance-RLCT area detection unit graph, and carrying out constant false-alarm detection on the distance-RLCT area detection unit graph; and (5) evaluating micro-motion feature parameters, and outputting motion trace points. The sea surface micro-motion target Radon-linear contact transformation long-time phase-coherent accumulation detecting method applies sea surface target micro-motion features to target detection, meanwhile utilizes range and phase information of the echo to perform long-time phase-coherent accumulation to carry out long-time phase-coherent accumulation, effectively compensates distance and the Doppler migration, improves accumulation gain, has the capacity to detect a micro-motion target in strong clutter, can obtain the motion trace points of the target, and has popularization application value.

The invention provides a video classification method and apparatus, a computer device and a readable medium. The method comprises the steps of obtaining multiple frames of pictures from a target video, identifying each frame of picture according to a pre-trained picture classification model, and predicting a probability of each preset type corresponding to each frame of picture; and according to the probability of each preset type corresponding to each frame of picture, obtaining a target type of the target video. According to the technical scheme, the technical problem of relatively low accuracy of classifying video contents directly based on a video title and a video reply in the prior art can be solved; and by predicting the probability of each preset type corresponding to each frame of picture in the target video based on the pre-trained picture classification model, the target type of the target video is obtained, so that the target video classification accuracy can be effectively improved.

A system and method for automatically generating URLs which reference one or more data sources or data targets connected to a computer. The data sources/targets may be of any type including files, http servers, ftp servers, hardware devices, etc. The generated URLs may be used together with a system such as the Data Socket system to seamlessly read data from disparate data source types or write data to disparate data target types without having to manually perform configuration, connection operations, data format conversion, etc. The URLs are generated automatically and contain all the information necessary for accessing the data source/target. This greatly abstracts a user or application from the I/O process.

The invention discloses an automobile active anti-collision early warning system based on millimeter-wave radars. The automobile active anti-collision early warning system mainly comprises a signal acquiring and processing module I, an automobile information acquiring module II, a judgment recognition module III, a man-machine interaction module IV and an execution system module V. The signal acquiring and processing module I utilizes the millimeter-wave radars I-1 to obtain the speed, a distance and an azimuth angle of a target X relative to an automobile Z and outputs the information processed by a processor I-2. The automobile information acquiring module II utilizes an ABS wheel speed sensor II-1 to obtain the automobile speed and road adhesion coefficient of the automobile Z and outputs the information processed by an auxiliary II-2. The judgment recognition module III judges the type of the target X, transmits the target type to the man-machine interaction module IV and transmits a control strategy to the execution system module V. By taking corresponding avoidance control measures, the automobile active anti-collision early warning system can remind a drive to take the measures so as to ensure safe and comfortable driving in advance of enough time, effectively recognize a dangerous target, remind the driver and improve the system safety.

The invention relates to a target positioning recognition system of an underground robot. The target positioning recognition system comprises a buoy subsystem, a sonar subsystem, a target recognition subsystem and a control subsystem. The buoy subsystem suspends on the water face above the underground robot. The target subsystem, the sonar subsystem and the control subsystem are arranged on the underground robot. The buoy subsystem is used for determining the buoy GPS position. The sonar subsystem is used for determining the buoy relative position of a buoy relative to the underground robot. The target recognition subsystem is used for determining the underwater target type and the target relative distance of an underground target relative to the underground robot. The control subsystem is used for determining the underground robot GPS position and the underground target GPS position according to the buoy GPS position, the buoy relative position and the target relative position. By means of the target positioning recognition system, the GPS position of the underground robot and the GPS position of the underground target can be accurately determined in an effective mode under the situation that underground GPS signals cannot be wirelessly transmitted, and meanwhile the type of the underground robot can be recognized.

A target type thrust reverser is provided for reversing the thrust of jet engines, particularly on aircraft. The thrust reverser preferably has a plurality of doors that occupy a stowed position about the nozzle of the jet engine until deployed. In the stowed position, the doors are out of the exhaust stream of the jet engine. The doors are mounted to pivot joints on the rear portions of the doors. To deploy the thrust reverser, actuators connected to the pivot joints cause the pivot joints to translate linearly aft, and link rods attached to the forward portions of the doors cause the doors to pivot about the sliding pivot joints. In this manner, the distance between the doors and the engine exhaust nozzle is minimized during deployment, which enables significant weight savings due to reduced loads and fewer parts. A novel lock is also provided for simultaneously locking the thrust reverser doors and the actuators.

The invention discloses a method and a device for identifying target types. The method comprises steps that a sound signal of a target is acquired through a planar sound transmission array, the target includes any one of a low-altitude aircraft, a light aircraft, a power delta wing, a cruise missile and a power paraglider, the direction of the sound signal is evaluated, spatial filtering processing on the sound signal after direction evaluation is carried out, after spatial filtering processing, characteristics of the sound signal, characteristic vectors corresponding to the characteristics of the sound signal and characteristics of the target are extracted, type confidence of the target is determined according to the characteristics of the sound signal, the characteristic vectors of the sound signal and the characteristics of the target, the type confidence is identified through the fusion identification technology, and thereby the type of the target is determined. Through the method, type identification accuracy of the target is improved.

The invention provides a method and device for accessing Internet of things equipment into an Internet of things platform and a server. The Internet of things platform comprises a connection adaptingsystem; and the connection adapting system comprises a connection public module and multiple type adapting modules respectively corresponding to different equipment types. The method comprises the following steps: when no brand model adapting module corresponding to a target brand model exists in the connection adapting system, acquiring brand model configuration data based on a target configuration template, wherein the target configuration template is a configuration template matched with a target equipment type; and establishing a brand model adapting module connected with a target type adapting module and corresponding to the target brand model in the connection adapting system on the basis of the brand model configuration data, so as to access the Internet of things equipment into theInternet of things platform by connecting the brand model adapting module. The method can realize the Internet of things platform compatible with all types of Internet of things equipment.

The invention provides a method and device for allocating operation and maintenance work orders. Among them, the method includes: determining the fault type of the faulty device based on the alarm information generated by the equipment in the site, wherein the faulty device is a faulty device in the site; The target operation and maintenance object matching the fault type; assign the operation and maintenance work order of the faulty device to the target operation and maintenance object, wherein the operation and maintenance work order is used to instruct the target operation and maintenance object to maintain the faulty device. The invention solves the technical problem of low operation and maintenance efficiency in the related art.

Apparatuses, methods and programs store a plurality of target types, each target type having recognition attribute information indicating the likelihood that the target type will be recognized based on a type of road or a type of area. The apparatuses, methods and programs acquire vehicle position information indicating a current position of a vehicle, acquire image information in the vicinity of the vehicle, and acquire road attribute information about a road on which the vehicle is traveling, the road attribute information acquired based on the vehicle position information. The apparatuses, methods and programs determine a target type defining a target for image recognition within the image information, the target type being determined based on the acquired road attribute information and the stored recognition attribute information, and perform image recognition of a target feature included in the image information, the target feature being of the determined target type.

Embodiments of the invention disclose a vehicle driving simulation method, a vehicle driving simulation device, an electronic device, a system, a program and a medium. The method comprises the following steps: generating a simulated driving environment of a simulated vehicle according to configuration information of driving environment parameters; generating a risky area according to the type, size and movement velocity of a target in the simulated driving environment, and dynamically updating the risky area according to the change of the movement state of the movement target in the simulateddriving environment; according to the control quantity over the simulated driving environment and the simulated vehicle, realizing the simulated driving of the simulated vehicle, and carrying out collision detection on the simulated vehicle based on the risky area; and by responding to the detection for the collision of the simulated vehicle, finishing the simulated driving of the simulated vehicle. With the embodiments of the invention, the simulated driving behavior is much closer to the real scene.

The invention discloses an image enhancement method. The image enhancement method comprises steps that binary processing on original images is carried out, and the original images are segmented into background type images and target type images; according to the image types after segmentation, calculation processing on the images after segmentation is carried out through employing a corresponding scale retinex algorithm; double-side filtering processing on brightness images contained in the images after calculation processing is carried out, wavelet de-noising processing on reflection images contained in the images after calculation processing is carried out; the images after double-side filtering processing and the images after wavelet de-noising processing are merged, gamma correction is further carried out, and the images after enhancement are acquired. Two-scale transform is carried out, and reduction of convolutional calculation amount is facilitated; double-side filtering processing on the brightness images is carried out, wavelet de-noising processing on the reflection images is carried out, noise removal for the images is facilitated, and image detail loss is relatively small; gamma correction for the images after synthesis is carried out, and global image enhancement is facilitated.

The invention belongs to the technical field of photoelectric tracking, precision measurement, navigation and location and particularly relates to a target tracking method based on intelligent IMM Kalman filter feedforward compensation and a system. A neural network intelligent target recognizer recognizes the type of a tracked target, state equation parameters of online operation of an improved intelligent IMM Kalman filter are adjusted, so that the filter is endowed with an adaptive characteristic, an online operating state model equation is matched with the tracked target, multiple filtersbased on different models are used for parallel work, speed and acceleration of the maneuvering target are estimated in real time accurately, estimation accuracy higher than that of traditional filtering is obtained, and a feedforward compensation controller constructed on the basis can obtain higher feedforward compensation precision and control performance.