76results about How to "Accurate targeting" patented technology

A method for accurately positioning the acetabular cup in a minimally invasive total hip arthroplasty (THA), comprising the steps of (a) placing the cup in roughly the correct position in the acetabulum using a acetabular component placement tool, (b) taking a first abduction reading and a first anteversion reading using a gyroscopic positioning unit aligned with the acetabular component placement tool, (c) taking an image of at least a portion of the cup using a radiography unit, (d) using the image to determine the actual orientation of the cup and required position of the cup to properly orient the acetabular component, (e) incrementally altering the position of the cup by using a striking tool, (f) taking new abduction and anteversion readings using the gyroscopic unit to determine the relative movement of the cup caused by the tapping, and (g) repeating any of the steps as necessary.

The invention relates to a target positioning method and a target positioning device for airborne equipment. The device comprises a signal processing unit, a laser distance measuring instrument, a thermal infrared imager, a visible light camera, a combined inertial navigation module and an angle meter, wherein output signals of the laser distance measuring instrument, the thermal infrared imager, the visible light camera, the combined inertial navigation module and the angle meter are all connected to the signal processing unit, and the signal processing unit is used for solving positional information of a target according to uploaded signals. Aerial carrier global positioning system (GPS) information, aerial carrier gesture information, angular deviation of observation equipment relative to an aerial carrier and distance measuring information of the target by the observation equipment are integrated to realize accurate positioning of longitude and latitude height of the target. The advantage of a multi-sensor is combined, a multi-dimensional spatial alternation principle is utilized so as to rapidly and effectively realize accurate target positioning in a non-contact and high-adaptability way, and the target positioning method can be applied to fields such as forest fire prevention, personnel rapid search and rescue and the like, and has wide technological and market prospects.

The invention provides an unmanned aerial vehicle target positioning method and system. The method comprises the steps that an image feature database of a satellite remote sensing image is established; an unmanned aerial vehicle reconnaissance image is acquired and preprocessed; multi-scale image feature extracting is carried out on the preprocessed unmanned aerial vehicle reconnaissance image; an image retrieval similarity measurement model based on granular computing is established, and matching computing is carried out on the extracted image features and image features in the image feature database according to the image retrieval similarity measurement model to acquire a similarity measurement value; and the geographic position corresponding to the satellite remote sensing image with the similarity measurement value satisfying a preset threshold value is determined as the unmanned aerial vehicle target position. The method and system can accurately position an unmanned aerial vehicle target.

The invention relates to the field of wireless communication, in particular to an acquisition and processing system of embedded global system for mobile communications for railway (GSM-R) network interference signals based on multiprocessors. The system comprises a processor, a power module, an analog to digital (AD) data acquisition module, a data and program storage module, a remote data transmission module and a data backup module, the AD data acquisition module coverts collected intermediate frequency analog signals of uplink frequency ranges and downlink frequency ranges of a GSM-R into digital signals to be sent to a digital signal processor (DSP) core in the processor, the DSP core operates and processes to store the digital signals in double data rate 2 (DDR2) chip in the data and program storage module, simultaneously the serial advanced technology attachment (SATA) connector stores the digital signals in a high-capacity rigid disk, the AD data acquisition module starts the intermediate frequency analog signals in an abnormal signal-collected frequency range when data acquisition is abnormal, and the AD data acquisition module is communicated with an upper computer through an Ethernet transceiver.

The invention discloses a casting surface defect identification method based on a deep convolutional neural network. The method comprises the following steps: 1, collecting a casting surface defect image, marking the image, and establishing a data set of common casting surface defects; 2, constructing a deep convolutional neural network defect recognition model; 3, constructing a network loss function; 4, dividing the data set into a training set and a test set, and training the defect recognition network by using the training set; 5, inputting the test image into the trained network to identify the position, the type and the size of the defect. According to the invention, the recognition precision and recognition performance of the casting surface defects are improved, and the online, intelligent and automatic development of casting quality detection is promoted.

The invention discloses a multi-terminal fusion cooperative command system based on augmented reality in the military field. The system comprises ten parts of user authority management, battlefield information source access, a command and control system integrated framework, external content loading, virtual battlefield space modeling, virtual battlefield space display, collaborative interaction sharing support, human-human image collaboration, human-target collaboration and human-target interaction. Aimed at problems that a traditional remote command cooperation mode in the existing militaryfield is low in efficiency and poor in real-time performance, cannot achieve visual communication and seriously depends on a mouse and a keyboard, a command environment is provided for all levels of commanders. A plurality of commanders in different places can be ensured to perform various command links such as situation plotting, simulation drilling, intelligence collection and wanted situation study and judgment based on the same virtual battlefield space in combination with the introduction of various information resources in the battlefield by implementing efficient human-human cooperation / character interaction, so that the cooperative command efficiency is greatly improved.

The invention belongs to the technical field of image detection and relates to a method for tracing a covariance matrix based on grayscale restraint, which comprises the following steps: selecting a rectangular area of a image as a target model to trace; calculating a grayscale value of each point of the target model; extracting the characteristic vector of each point of the target model; calculating the covariance matrix of the target model; after the length and the width of a subsequent frame are respectively amplified by taking the target model as a center, obtaining a tracing window, selecting a candidate target and checking whether the candidate target meets the grayscale restraint or not; and calculating a difference value of the covariance matrix of the candidate target which meetsthe grayscale restraint and the target model, wherein the candidate target area with the minimal difference value is the position of a traced target. The invention has the advantages of more accuratetarget positioning, higher tracing speed and higher instantaneity.

The invention discloses a convolutional neural network model-based violence and terrorism video detection method, and relates to computer vision and machine learning. The method comprises the following steps of 1) training a deep neural network model; and 2) detecting violence and terrorism videos online. By utilizing a low-level feature of a deep learning model combination, a more abstract high-level representation attribute or feature is formed to discover distributed feature representation of data. Video image feature descriptors with good description capabilities can be obtained through the model. The feature descriptors cover feature information, at all levels from low to high, of video images, so that the detection precision and recall rate of violence and terrorism videos are greatly improved and increased. A deep convolutional network is trained through a small amount of samples to obtain excellent detection performance. The detection precision of terrorism pictures reaches more than 99%, and the recall rate of the terrorism pictures reaches more than 98%. The detection precision of terrorism videos reaches 95%, and the recall rate of the terrorism videos reaches 99%. The training process is free from artificial participation, and massive data is generated automatically according to a small amount of the samples.

The invention relates to a method and device for locating a target in a virtual interactive scene. The method includes the steps of monitoring a triggering operation of a user for target locating andobtaining the position of a trigger point in a screen; according to the position of the trigger point and a plurality of adsorption points on a virtual target, determining the positions of the adsorption points, wherein the position of each adsorption point corresponds to one adsorption point; when the position of the trigger point stays in an adsorption range determined by the position of any adsorption point, performing adsorption point weight value operation on the adsorption points to obtain adsorption point weight values corresponding to the adsorption points; moving the trigger point tothe position of the adsorption point corresponding to the maximum adsorption point weight value. By adopting the method and the device, the flexibility of target locating in the virtual interactive scene is effectively improved.

The invention discloses an indoor radio tomography imaging enhancement method capable of deleting an interfering link based on motion prediction, and belongs to the technical fields of target detecting and tracking in a radio network. The method comprises the following steps of finding out an intersection set formed by shadow fading links by virtue of clusters; taking the center of the set as estimation of a target position at an initial moment, and utilizing a target motion model and the positioning result of the last moment to obtain estimation of the target position at the current moment through Kalman filtering, so as to delete the interfering link which is relatively far away from the target position; finally carrying out RTI (radio tomography imaging) through the shadow fading links, and obtaining the current moment position by using Kalman updating on the basis of the RTI. The effects of non-shade fading links are eliminated based on the motion prediction; false targets can be effectively removed; the fading effect caused by presence of the target is relatively accurately highlighted, thus target positioning is relatively accurate in a multipath environment; the imaging quality is improved; relatively excellent dynamic tracking is achieved.

The invention provides a target tracking method based on online state learning and estimation and belongs to the computer vision, computer graphics and image technical field. According to the method, a target positioning and state estimation network is obtained; and the target positioning and state estimation network is composed of a feature extraction network and a regression network, wherein the feature extraction network is a pre-trained AlexNet network, and the regression network is a recurrent neural network (RNN). In an initial network training process, an initial training set and a stochastic gradient descent method are utilized to train the target positioning and state estimation network, after being trained, the target positioning and state estimation network obtains an initial ability to perform target positioning and state estimation. In a tracking process, the target positioning and state estimation network performs forward processing on an inputted image and directly outputs target related information corresponding to the image, wherein obtained target probability and state information decides whether the network to perform online learning, and target position and size information can realize the localization of a target, and therefore, the tracking of the target can be realized.

The invention discloses a shadow fading assistance carry-free equipment positioning method, and belongs to the technical field of wireless positioning in a wireless network. Wire nodes surrounding a monitoring region build a fingerprint database in an offline stage through the RSS (received signal strength) change of a measuring object in each training position; in the online stage, the shadow fading link detection is firstly performed; then, the positioning algorithm is selected according to different conditions of the shadow fading region intersection set. When an objective does not detect the shadow link in the position, a conventional fingerprint recognition method is still used; when the shadow fading link exists, the mode matching is only performed in a shadow fading region; when the shadow fading links are many, for example, the objective is positioned in the center of the monitoring region, the center of the intersection region is taken to be used as the objective position estimation. The method has the advantages that the shadow fading and the fingerprint recognition are combined; the shadow fading link is effectively utilized for assisting the fingerprint recognition, so that the target positioning is more precise.

The invention relates to a miniature large-field-of-view laser fuze transmitting-receiving optical system. The miniature large-field-of-view laser fuze transmitting-receiving optical system is composed of three transmitting optical systems and three receiving optical systems; the three transmitting optical systems and the three receiving optical systems are arranged in series in a circumferential direction; each transmitting optical system and each receiving optical system are arranged adjacently and alternately; an included angle between the central optical axes of two adjacent transmitting optical systems is 120 degrees; an included angle between the central optical axes of two adjacent receiving optical systems is 120 degrees; in an arc vector direction, each transmitting optical system uses a 120-degree transmitting field of view, and each receiving optical system uses a 120-degree transmitting field of view; and in a meridian direction, each transmitting optical system uses a 1-degree transmitting field of view, and each receiving optical system uses a 2-to-8-degree receiving field of view. With the miniature large-field-of-view laser fuze transmitting-receiving optical system of the invention adopted, laser fuze can perform large-field-of-view, circumferential and leakage-free detection on any target, and a requirement for the miniaturization of the laser fuze is satisfied. The miniature large-field-of-view laser fuze transmitting-receiving optical system has the advantages of accurate target positioning, high detection precision and strong anti-interference capability.

The invention belongs to the field of human-machine interaction and particularly relates to a multi-channel accurate target positioning method for touch equipment. According to the multi-channel accurate target positioning method, touch input and voice input are used for positioning the target, wherein the touch input is used for providing a possible target object candidate set; the voice input is also used for providing a possible target object candidate set; and the possible target object candidate sets are positioned to an accurate target object by a multi-channel fusion algorithm. According to the multi-channel accurate target positioning method disclosed by the invention, two natural input channels, such as the touch input and the voice input as well as two natural interaction manners are supported; more accurate target positioning can be carried out; and the efficiency of positioning relevant tasks on the touch equipment is increased.

The invention discloses a multiple carrier frequency MISO radar target locating method based on priori knowledge and mainly aims at solving the problem that in the prior art, the target distance speed coupling is serious so that target speed can not be obtained. The method comprises the following steps: (1) receiving target return signals; (2) pre-processing; (3) confirming a target roughly-measured distance; (4) integrating pulses and processing apertures; (5) confirming the target distance; (6) locating targets. The multiple carrier frequency MISO radar target locating method based on the priori knowledge utilizes the priori knowledge to confirm a target locating angle and the target distance so as to avoid the problem that target speed can not be solved in a single frequency modulation cycle and achieve the purpose of precisely solving the target distance. The multiple carrier frequency MISO radar target locating method is applied to the condition that in the actual application, requirements for signal data processing efficiency is high, and can be used by multiple carrier frequency MISO radar to locate aerial targets.

The invention discloses a weighted sparse cooperative model-based target tracking method. The method comprises the following steps of: 1, selecting a tracking target from the first frame of image data of a video sequence; 2, initializing any frame f=1; 3, processing the fth frame of image data in the video sequence by utilizing a particle filter algorithm so as to obtain a plurality of target candidate boxes; 4, processing the target candidate boxes in the fth frame of image data by utilizing a weighted sparse representation-based discriminant algorithm so as to obtain discriminant scores of the target candidate boxes; processing the target candidate boxes in the fth frame of image data by utilizing a weighted sparse representation-based generative algorithm so as to obtain generative scores of the target candidate boxes; and 6, obtaining final scores of the candidate boxes, comparing the scores of all the candidate boxes to find a candidate box corresponding to a maximum value, and taking the found candidate box as a tracking result. According to the method, real-time movement estimation and positioning can be carried out on moving targets in a video sequence, so that the stable tracking for the moving targets is realized.

The invention discloses hydroxyapatite-coated magnetic drug-loaded nanoparticles as well as a preparation method and application thereof in preparation of osteosarcoma phototherapy drugs. The hydroxyapatite-coated magnetic drug-loaded nanoparticle comprises a polydopamine magnetic nanoparticle core, and the outer side of the polydopamine magnetic nanoparticle core is sequentially coated with a drug layer and a hydroxyapatite layer; the hydroxyapatite layer is adopted to completely wrap the drug layer, loaded drugs such as a photosensitizer cannot be released in advance, and the drug targetingeffect of the magnetic drug-loaded nanoparticles is accurate and effective; the photosensitizer and other drugs do not generate optical toxic and side effects on normal tissues; hydroxyapatite can behydrolyzed in the tumor environment with the pH being about 5, then a large amount of photosensitizer is released and retained in the tumor, it is more suitable for in-vivo application. Hydroxyapatitecan promote osteoblast proliferation, kill tumor cells, sterilize and promote bone tissue repair at the same time, and three purposes are achieved at a time.

The present invention discloses a target tracking method based on super-resolution reconstruction. The method comprises the steps of firstly utilizing a nonlinear iteration algorithm to carry out the super-resolution reconstruction on a low-resolution image of a target image sequence; secondly, standardizing the image after the super-resolution reconstruction and determining a desired motion target area; finally, combining a variable template method to detect and track a target in the desired motion target area. The target tracking method based on the super-resolution reconstruction of the present invention better solves the detection difficulty caused by the deformation in the target motion, and is good in detection and tracking effects.

The invention provides a target tracking method based on spatial offset learning, which relates to the technical field of computer vision. The target object to be tracked is selected and determined, and the target selection process is automatically extracted or manually specified by the moving target detection method. Spatial migration learning network includes four parts: image data extraction, depth neural network, multi-layer perceptron MLP and spatial migration output. In real-time processing, the video image collected by camera and stored in storage area is extracted as the input image tobe tracked. In the case of offline processing, the captured video files are decomposed into image sequences composed of multiple frames. Short-time tracking adopts particle filter, in which a particle represents a possible target image block, and the corresponding target attention area ROI is used as the online training set of the spatial offset learning network, and the stochastic gradient descent (SGD) method is used to online train the spatial offset learning network and update the network parameters. Target Location and updating are carried out.

The invention provides an underwater image recognition tracking system and method, and relates to the technical field of computer vision. The system comprises an affine transformation module, a candidate box generation module, a tracking module and a training module. The method comprises the following steps: receiving an underwater video image sequence, and performing affine transformation on images in the received image sequence to mark an affine frame on a current image frame; on the basis of the affine box, carrying out affine transformation by using a random walk model to form M candidateboxes; inputting candidate image regions determined by the M candidate boxes into a deep convolutional neural network to obtain M scores, wherein M is an integer greater than 1; and determining the candidate box with the highest score as an object to be identified, and marking out the candidate box in the current frame as an identified and tracked object.

The invention discloses a large vibro-stand multi-bearing embedded state monitoring system. The system is composed of a plurality of acceleration sensors, a plurality of temperature sensors, a sensorresponse data collection front end, a signal analysis processing platform, and signal real-time display equipment; the acceleration sensors and the temperature sensors are used for monitoring acceleration responses and temperature signals of various bearing bodies of the large vibro-stand system in real time; the signal analysis processing platform uploads and saves the signals of various sensors,performing postprocessing and analysis computation on various sensor signals to obtain stat indexes; and the signal real-time display equipment displays the real-time response data of various sensorsand the postprocessing results. The monitoring system disclosed by the invention can master a working state of the large vibro-stand in real time, thereby discovering the abnormity in time; the system has a good human-machine interaction interface and is reasonable in design, convenient to realize, convenient to operate, high in monitoring precision, wide in range, complete in function and strongpracticability, and especially suitable for the large vibro-stand with hundreds of T-shaped sliding bearings.

The invention belongs to the technical field of gene editing, and in particular relates to a CRISPR / Sa-SauriCas9 gene editing system and application thereof. The gene editing system of the invention is a complex formed by Sa-SauriCas9 protein and sgRNA, and can accurately target DNA sequences and generate cleavage to cause double-strand breaks to DNA. The gene editing is to perform gene editing incells or in vitro; Sa-SauriCas9 is a fusion protein, and the PAM recognition domain (PAM interacting, PI) of SaCas9 is replaced with the PAM recognition domain (SauriCas9-PI) of SauriCas9. The Sa-SauriCas9 protein is small and is 1056 amino acids, and the recognized PAM sequence is simple. The Sa-SauriCas9 protein has the amino acid sequence as shown in SEQ ID NO: 1, and the sgRNA has the nucleotide sequence as shown in SEQ ID NO: 2. The invention has wide application prospects in the field of gene editing.

The invention is suitable for the technical field of structural health monitoring, and provides a structural vibration displacement measurement method and device based on red circular target image processing, and the method comprises the steps of carrying out the video sequence decomposition of an obtained red circular target video, and obtaining each frame of image; separating the red circular target in each frame of image from the black square target to obtain a target pixel value; extracting the edge of the red circular target to obtain an edge pixel value; obtaining a circle center initialvalue and a radius initial value of the red circular target according to the target pixel value; filtering the edge pixel value according to the circle center initial value and the radius initial value; calculating a target circle center value and a target radius value of the red circular target according to the filtered edge pixel value; according to the target circle center value and the targetradius value obtained through calculation of the multiple frames of images and the measured actual radius value of the red circular target, obtaining the actual displacement change value of the red circular target through calculation, and therefore target positioning can be rapidly and accurately conducted.

The invention discloses a method, a device and a system for target positioning. The method comprises the following steps: cluster nodes of a positioning system obtain measurement data corresponding to at least one target device, wherein a positioning vector reflecting the position of the at least one target device in the positioning system has sparsity; decorrelation processing is performed on the measurement data; reported data is sent to the processing center of the positioning system, so that the processing center determines the position of the at least one target device in the positioning system according to the reported data from the at least two cluster nodes and the low rank characteristic of positioning matrixes, wherein the reported data comprises the measurement data subjected to decorrelation processing, and the positioning matrixes are column vector matrixes of which the column number is the same as the quantity of the cluster nodes; the column vectors of the positioning matrixes are positioning vectors; and the independent sparsity of the single positioning vector is converted into low rank characteristic of the positioning matrixes comprising the positioning vectors of which the column number is the same as the quantity of the cluster nodes, so that the accurate target positioning is realized.

The invention discloses a text target detection method and system based on a deep network, and the method comprises the steps: selecting an original image, and extracting a feature map of the originalimage; judging whether the anchor point frame of the feature map is a foreground or a background, and correcting the anchor point frame by using a frame regression device to obtain a proposal area; collecting the input feature map and the proposal area, and extracting the feature map of the proposal area; segmenting the feature map of the proposal area into a left upper corner point feature map,a right upper corner point feature map, a right lower corner point feature map and a left lower corner point feature map of the text target; and performing regression of corresponding angular point coordinates on each feature map to obtain a frame of the original picture.

The invention discloses a target positioning method for searching and exploring of an underwater robot. The method comprises the following steps: extracting and identifying A-KAZE feature points of anunderwater target sonar image; measuring and calculating the two-dimensional orientation of the underwater target relative to the underwater robot; measuring and calculating the elevation angle of the underwater robot; measuring and calculating the three-dimensional orientation of the underwater target relative to the underwater robot; and correcting the elevation angle of the underwater robot, and correcting the three-dimensional orientation of the underwater target point relative to the underwater robot. The invention discloses a target positioning method for searching and exploring of an underwater robot. The method is based on forward-looking sonar data. A deep convolutional neural network is used to automatically extract and identify underwater target feature points; the underwater target is accurately positioned in combination with the posture of the underwater robot, search personnel can conveniently perform refined detection on the position of the underwater target, reliability, high efficiency and intelligence of underwater search and exploration operation are realized, and the method is applied to the technical field of underwater target search and exploration.

The invention discloses aerial spraying equipment. The aerial spraying equipment comprises a liquid storage box and a spraying pipe which are used for being carried on a light aircraft, wherein the bottom end of the liquid storage box communicates with the spraying pipe through a liquid discharge pipe; the spraying pipe is connected with a plurality of electrostatic nozzles, and electrostatic generators are arranged outside the electrostatic nozzles in a cooperative mode; the electrostatic generators are used for being connected with a control system arranged in the light aircraft; and each electrostatic nozzle comprises an electrostatic spraying bracket, a nozzle body and a high voltage electrode, wherein the nozzle body is fixedly connected with the electrostatic spraying bracket, the high voltage electrode is fixed with the electrostatic spraying bracket through a self-locking nut, and the high voltage electrode is connected with each electrostatic generator. The aerial spraying equipment disclosed by the invention has the advantages that the spraying target is more accurate, the pesticide spraying efficiency is high, the effective dosage can be increased, and waste and environmental pollution can be avoided.

The invention provides a high-accuracy intelligent identification target tracking system and method for a security camera, and the system and method solve a problem of class imbalance through the cascading of a series of RPN (Region Proposal Network), achieve the full mining of cross-layer features, and achieve a stable visual tracking function. According to the invention, a new multi-stage tracking framework, namely the twin cascaded RPN, is proposed for the first time, a tracker based on the twin cascaded RPN is achieved, the recognition capability of the twin cascaded RPN for utilizing advanced semantic information and low-level spatial information is further improved, and through multi-step regression, the positioning is more accurate, and the distribution sequence of training samples is more balanced; the classifier of the RPN is more discriminative in sequence when more difficult interferents are distinguished, and target positioning is more accurate and has real-time performance under a complex background.