364 results about "Target capture" patented technology

The invention provides an unmanned aerial vehicle identification method and system for low-altitude security and protection. An optical compensation parameter set for unmanned aerial vehicle imaging optimization is generated in real time through a joint optimization model of an ambient light intensity change rate and a background motion vector field, and an original optical sequence in a target capture window is processed by using the parameter set. And constructing a time domain deconvolution kernel in combination with the motion characteristics of the unmanned aerial vehicle to generate an enhanced optical image resistant to motion blur. Non-linear weighted fusion is carried out through a disturbance intensity evaluation function, and a confrontation disturbance feature mask is formed. The enhanced optical image and the confrontation disturbance feature mask are subjected to airspace superposition operation, a multi-scale residual network is adopted to carry out target confidence estimation on the superposed image, an unmanned aerial vehicle recognition result is generated, and the unmanned aerial vehicle recognition accuracy and the anti-interference capacity in the low-altitude complex environment are remarkably improved through the technical scheme provided by the invention.

The invention provides a moving target capturing method and system based on AI multi-view vision, and relates to the technical field of data processing.The method comprises the steps that camera parameters in all vision systems are calibrated, and the camera parameters comprise a camera internal parameter matrix and a camera external parameter matrix; extracting 2D feature points of the moving target through each calibrated visual system; reconstructing each 2D feature point by combining a multi-view geometric reconstruction algorithm and a least square method according to the calibrated camera parameters to obtain 3D feature points; constraints are added to the 3D feature points, and the constraints comprise spatial constraints, direction vector constraints and time sequence constraints; reconstructing the 3D feature points added with the constraints through a multi-view geometric analysis algorithm to obtain three-dimensional motion information of the moving target; and outputting the three-dimensional motion information of the moving target to complete the capture of the moving target. And non-contact high-precision capture of the moving target can be realized.

The invention discloses an unmanned aerial vehicle cluster collaborative target hunting method based on brain-like calculation, and belongs to the technical field of autonomous navigation robot and multi-robot coordination control, and the method comprises the steps: obtaining a multi-view image under a camera view, and obtaining the local pose and covariance of each unmanned aerial vehicle; obtaining a global consistent relative pose of the unmanned aerial vehicle cluster according to the local pose and the covariance of each unmanned aerial vehicle; performing state estimation on the target by adopting an unscented Kalman filtering algorithm to obtain target state information under a unified coordinate system; using a Bezier curve generation algorithm to obtain a target motion trail in a section of historical state; the method comprises the following steps: expanding an unmanned aerial vehicle cluster through a centroid extension method to obtain a surrounding queue of target surrounding, solving the minimum surrounding cost by adopting a Gaussian Newton method, and generating a surrounding position of each unmanned aerial vehicle; the motion trajectory of the unmanned aerial vehicle is generated by adopting mixed A * search, and a final surrounding trajectory sequence is generated, so that the accuracy of motion state estimation is remarkably improved.

The invention discloses a pumped storage power station safety construction monitoring method based on image recognition, and relates to the technical field of pumped storage power station safety construction monitoring, and the method comprises the steps: carrying out the intelligent monitoring of a construction site through a high-definition camera, an unmanned plane, and a multispectral imaging device, and optimizing the target capturing capability through an environment adaptive adjustment strategy. And in combination with a deep learning and behavior analysis model, identifying construction personnel, equipment and environment states, detecting behaviors such as no safety helmet wearing, illegal retention, equipment abnormity and the like, and performing real-time tracking and updating. Based on a risk level calculation model, a multi-level alarm mechanism is constructed, acousto-optic early warning, remote alarm and task scheduling are triggered, and construction safety is optimized in combination with cloud collaborative management. And self-supervised learning and cross-scene transfer learning are adopted, so that the recognition precision and the system adaptability in a complex environment are improved. According to the method, better effects are achieved in the aspects of real-time performance and accuracy.

The invention relates to the technical field of artificial intelligence, can be applied to business scenes such as agent autonomous decision making, financial science and technology and medical health, and discloses a task processing method and device based on visual attention enhancement, equipment and a medium. Visual hierarchical features are extracted, a double fovea attention module processes and fuses high-level visual features, a side suppression network obtains enhanced visual features, and a cross-modal fusion module generates fusion features by taking the enhanced visual features as query vectors and taking language components and action components as key and value vectors; and fusing the feature input decision network to generate target category and position information, generating feedback information based on actual label difference, and updating module parameters to complete a target task. According to the invention, through combination of a bionic vision mechanism and multi-modal attention fusion, the visual feature extraction and background suppression capability is improved, and the target capture efficiency and recognition precision in a complex scene can be improved.

A low-density liquid-phase chip for Gossypium hirsutum L. based on targeted capture sequencing and use thereof provided. The low-density chip includes 908 SNP loci, where among the 908 SNP loci, 329 loci are significantly related to important agronomic traits, such as fiber quality, yield, and disease resistance of Gossypium hirsutum L., 14 loci are common plant transgenic detection loci, and 565 loci are other loci. This chip is adapted for use of detection of transgenic components in Gossypium hirsutum L. varieties, resource evaluation and lineage identification, seed purity identification, and genetic improvement of major agronomic traits of Gossypium hirsutum L. varieties.

The invention relates to the technical field of crop genetic breeding and plant molecular design breeding, in particular to a pakchoi whole genome liquid phase chip and application. The pakchoi breeding gene chip disclosed by the invention comprises a target capture probe which is designed by taking 1000 core SNP (Single Nucleotide Polymorphism) molecular markers positioned on 40KSNP sites on a pakchoi reference genome version V4.0 as templates. The pakchoi whole genome liquid phase chip provided by the invention comprises a 40K site targeted capture probe mixed solution and a hybridization capture reagent. The pakchoi 40KSNP targeted capture probe can rapidly and effectively track genetic materials of pakchoi, and is suitable for different application scenes such as accurate identification of pakchoi germplasm resources, genetic relationship analysis between materials, variety authenticity identification, QTL gene positioning and mining, molecular marker-assisted breeding, variety improvement and whole genome selective breeding application. Therefore, the method has good application prospect and important social value.

The image processing device 1X mainly includes a three-dimensional reconstruction means 31X, a matching means 32X, a complement means 33X, and a display control means 34X. The three-dimensional reconstruction means generates, based on endoscopic images of an examination target captured by a photographing unit provided in an endoscope, reconstruction data obtained by three-dimensionally reconstructing the examination target. The matching means performs matching between a three-dimensional model of the examination target and the reconstruction data. The complement means generates, based on a result of the matching, complemented reconstruction data which is the reconstruction data complemented with the three-dimensional model. The display control means displays the complemented reconstruction data on a display device.

The invention relates to the technical field of target recognition and capture, in particular to an unmanned aerial vehicle for power transmission line inspection and a target capture method thereof, and the method comprises the steps: determining a second interference degree through analyzing the dispersion degree and energy distribution of the change trend of pixel values in pixel point neighborhoods in different regions in an inspection graph, so as to obtain a stable frame; on the basis of the difference between the discrete degrees of pixel values of all pixel points in neighborhoods of different edge points in each stable frame and the curvature mean value of all the edge points on each edge contour, and in combination with the difference between the optical flow vectors of all the edge points in each stable frame, determining a characteristic value so as to obtain a positioning map; and the foreign matters on the power transmission line are grabbed. The invention aims to solve the interference of the vibration of the unmanned aerial vehicle on the visual feature extraction of the foreign matters on the power transmission line, reduce the false detection and missing detection risks of vegetation tree shielding on the foreign matters on the power transmission line, and improve the positioning precision in the target grabbing process.

The invention discloses an SNP (Single Nucleotide Polymorphism) and InDel site combination for identifying variety of Inner Mongolia cattle, a 5K targeted capture gene chip and application of the 5K targeted capture gene chip. The SNP and InDel site combination for identifying the variety of the Inner Mongolia cattle is firstly disclosed and comprises 5000 SNP and InDel sites, and the physical position information of a cattle reference genome with the version number of AR-UCD1.2 is shown in a table 1. The invention further discloses a 5K targeting capture gene chip comprising the probe targeting the SNP and InDel site combination as well as a synthesis method and application of the 5K targeting capture gene chip. The 5K targeted capture gene chip completes genetic typing through a targeted capture probe by means of a high-throughput sequencing technology, has the characteristics of low cost, high efficiency and high accuracy, and can realize the variety identification function of Mongolian cattle, Sanhue cattle and Sunitt cattle by utilizing genotype data; important technical support and specific means are provided for identification and evaluation of Inner Mongolia cattle germplasm resources.

The invention relates to the field of industrial automation, in particular to a material grabbing method, device and equipment and a program product. The method comprises the following steps: collecting a first image comprising a material through a top camera; determining a target capture area of the first image through a preset first visual model; the mechanical arm is controlled to move to the target grabbing area, and a second image including the target grabbing area is collected through the wrist camera; determining an action sequence and a grabbing pose of the mechanical arm for grabbing the material in the second image through a preset second visual model; and the materials are grabbed according to the action sequence and the grabbing poses. According to the method, the reliability in a disordered stacking scene can be improved, even if the materials move in the grabbing process, the output action sequence can be adjusted in time through timely updated observation of the wrist camera, the grabbing success probability can be improved, and the method can effectively adapt to grabbing operation in a dynamic change scene.

The invention discloses a humanoid robot falling prediction method, device and equipment and a storage medium, and relates to the technical field of robot control, and the method comprises the steps: obtaining target state information and a target capture point position corresponding to a target humanoid robot in real time in a task execution process of the target humanoid robot; whether the target humanoid robot can reach the target capture point position or not is judged based on the target state information, if yes, the predicted falling time of the target humanoid robot is determined, and the target motion time of the target humanoid robot reaching the target capture point position is determined; and if the predicted fall-down time is greater than or equal to the target motion time, generating a ZMP sequence of the target humanoid robot through a model prediction control algorithm, and predicting whether the target humanoid robot falls down or not based on the ZMP sequence and the target capture point position so as to perform fall-down protection on the target humanoid robot. According to the method, the dynamicity and timeliness of fall prediction of the humanoid robot can be improved, and safety protection is carried out during fall.

The invention discloses a spread spectrum communication multi-target capturing-to-tracking fast channel matching method, device and system, and belongs to the field of wireless communication, and the method comprises the steps: S1, obtaining capturing and tracking parameters, and initializing channel matching parameters; s2, identifying the same target from capturing / tracking parameters based on a code load consistency criterion to perform channel matching; s3, channel allocation is carried out on capturing and tracking parameters of different targets based on the criterion of capturing after tracking; and S4, resolving and correcting the pseudo code phase in the channel matching parameters. The method solves the problem of channel matching between multi-target pseudo code capturing and code load tracking of a receiving terminal of a spread spectrum communication system, has the advantages of capturing and tracking dynamic decoupling, high code load matching precision and good channel processing timeliness, and can realize fast channel matching processing of multi-target pseudo code capturing and code load tracking of same-frequency same-code spread spectrum signals.

Techniques for optical lens characterization and calibration are described herein. In an example, a computer system receives a first image of a target captured by an image acquisition system having a camera and a lens and using first setting values in a setting space. The computer system inputs the first setting values and a first decode performance associated with image acquisition system having the first setting values for the barcode sets in the first image into a machine learning model and determines a representation of the setting space. The computer system inputs the machine learning model and one or more conditions into an algorithm and receives an output indicating second setting values in the setting space for the image acquisition system. The computer system sends the second setting values to the controller, which is configured to set the settings for the image acquisition system based on the second setting values.

The invention belongs to the technical field of spacecraft ground simulation testing, and discloses a high-precision space GNC full-physical simulation test platform which is composed of a three-dimensional motion execution system, a distributed real-time control system, a capture interaction simulation system, a safety protection system and a three-dimensional motion simulation and path planning system. By integrating a three-dimensional track type large-range motion execution system and a double-grating ruler closed-loop detection structure, the simulation motion range is remarkably expanded, the technical bottleneck that an existing platform is limited by a single axis, a two-dimensional sliding rail and limited motion amplitude is broken through, high-precision cooperation of a motion control system and a track system is achieved through the platform, and high-precision cooperation of the motion control system and the track system is achieved. The space high-dynamic motion simulation of a large-size and high-inertia complex target can be met, autonomous tracking of a target flight path, flexible capture of a flying claw and small multi-degree-of-freedom control verification are supported through integrated design of a capture interactive simulation system, and complex tasks such as space rendezvous and docking, target capture and micro-force control can be effectively simulated.

The embodiment of the invention discloses a cluster robot self-organization control method based on environment gradient and local interaction. According to the method, individual autonomous motion control is achieved through distributed sensing and behavior decision, a light intensity gradient field is constructed through environment interaction to achieve implicit communication, and by combining state analysis, cluster synchronization and firmware maintenance functions, a robot can self-organize to emerge macroscopic intelligent behaviors such as feedback adjustment and target capture under the condition of no centralized control. The corresponding system can comprise a robot individual, an external state analysis device, a synchronous beacon and an updating server, through cooperative work of all the modules, the problems that a traditional system is complex in communication and poor in expandability are effectively solved while individual simplicity is kept, and low-cost and high-robustness cluster cooperative control is achieved.

The invention discloses a dairy cow 40K medium-low density SNP chip based on targeted capture sequencing and application thereof. The invention provides a chip for detecting genotypes of dairy cows. The chip comprises a reagent for specifically detecting SNP (Single Nucleotide Polymorphism) sites; the SNP loci are 44373 SNP loci in a table 2; the position information of the 44373 SNP loci in the table 2 is determined by comparison based on a dairy cow genome reference sequence, and the dairy cow genome reference sequence is in the version of Bosula UMD3.1. Compared with the existing 30K, 126K and 150K chips, the 40K chip disclosed by the invention has consistent detection performance in the aspects of SNP (Single Nucleotide Polymorphism) detection rate, polymorphism detection, individual genotype detection rate, genetic evaluation accuracy and the like. The chip disclosed by the invention can be applied to cow whole genome selection and genetic defect identification, and has important significance on early breeding of replacement cows, cow core group establishment, seed selection and matching, acceleration of genetic progress and improvement of feed management efficiency.

The invention relates to a composite self-adaptive compliant control method for synchronously capturing a target by a space double-arm robot. The method comprises the following steps: firstly, establishing a dynamic model of the space double-arm robot, a target to be captured and a contact process; in order to ensure that the tail ends of the two arms simultaneously reach the surface of the to-be-captured target, a tail end path planning algorithm based on a B spline curve is designed; then, in order to reduce impact borne by the double-arm robot system in the capturing process, an impedance control strategy is designed according to contact force borne by the tail ends of the double mechanical arms; secondly, the resultant force borne by the target is decomposed into internal force and external force through the kinematics and geometrical relationship; and finally, designing a self-adaptive internal force tracking impedance control strategy in order to prevent capture failure or target damage caused by simultaneous target capture by two arms. Aiming at a task scene in which the space double-arm robot synchronously captures the target, the problem that the space double-arm robot system is subjected to too large rigid impact and too large / too small clamping force, so that the target is damaged or clamping fails can be solved.

The invention discloses a shooting method and wearable equipment. The shooting method is applied to the wearable equipment with a camera. The method comprises the steps that a preview frame corresponding to a shooting target is obtained, and the preview frame is single-frame image data in a target preview flow collected by a camera and is not displayed in the collection process; fusing the preset visual features of the preview frame to obtain a decision score of the preview frame; the preset visual features comprise at least two items of a target length-width ratio, an edge density ratio and an image scene category; and analyzing the decision score based on a dynamic threshold strategy, and adjusting the shooting direction of the shooting target. In a scene that a user cannot preview a shot picture, multi-dimensional visual feature analysis is carried out on a preview frame in a real-time preview stream shot by the wearable device, after visual features of all dimensions are fused, a decision score is obtained, then the decision score is analyzed through a dynamic threshold strategy, and in a scene that the shot picture cannot be previewed, the preview frame in the real-time preview stream shot by the wearable device is subjected to multi-dimensional visual feature analysis. And the shooting direction of the shooting target is adjusted.

The application provides a variable configuration spacecraft system and method for non-cooperative target capture, and belongs to the field of deep space exploration. The method comprises the following steps: 1) a rigid combination flies to the vicinity of the target; 2) the combination is spun to be expanded into a ring belt; 3) the ring belt envelopes the target; and 4) the ring belt encloses the target. Large space debris and small celestial bodies both belong to non-cooperative targets, and cannot provide special interfaces for assisting capture. It is extremely difficult to develop general control mechanisms and methods to realize reliable capture and stable connection of various targets. In order to solve the problem, the application utilizes the inter-satellite connection to expand a plurality of controllable satellites into a ring-shaped multi-body tether satellite formation system, and carries out ring belt enveloping and then enclosing the tether to realize capture of a non-cooperative target with large scale, complex geometric shape, great uncertainty of motion state and inertia information according to a predetermined trajectory. The application has the advantages of high applicability, flexibility, high intelligence, reusability and small size.

The invention discloses an SNP (Single Nucleotide Polymorphism) molecular marker combination for predicting the shape of Chinese rose petals and application of the SNP molecular marker combination. Based on whole genome re-sequencing data of 358 Chinese rose germplasm resources, significant SNP sites with the length (2882), the width (2360) and the length-width ratio (2581) of the petals are screened out through multi-model GWAS analysis, and a high-precision prediction model (BRR model R2gt; 0.651) of the optical fiber. A liquid phase probe is synthesized on the basis of an Agilent SureDesign platform, a targeted capture chip is integrated, the dynamic development trajectory of petals can be predicted through genotyping in the seedling stage, and the longitudinal growth potential and the transverse expansion trend are quantified. According to the method, the technical bottleneck of traditional static measurement is broken through, target strains are accurately screened within 6-8 months after sowing, the breeding period is shortened to 2-3 years, and meanwhile the field planting and manual screening cost is reduced.

A method for characterizing and calibrating a helmet-mounted digital visual augmentation device for a user, wherein the method comprises: with cameras mounted on the helmet, acquiring stereo images of a calibration target; calibrating a position of the user's eyes; with a projection system, projecting structured-light patterns associated with the target on an AR / VR display; acquiring a 3D model of the AR / VR display and obtaining a shape of the AR / VR display; mapping between AR / VR display pixel coordinate and the position in 3D on the calibration target in front of the user's eyes where the pixel is perceived by the user; encoding deviations of the projection system based on the mapping, wherein the 3D model of the AR / VR display is in a same reference frame as the cameras mounted on the helmet; using at least one machine learning model to correct deviations between an image of the target as seen from a point of view of the user's eyes and an image of the target captured by the cameras; and generating a digital twin of the digital visual augmentation device.

The invention provides methods for non-invasive prenatal testing that allow for detecting risk of chromosomal and subchromosomal abnormalities, including but not limited to aneuploidies, microdeletions and microduplications, insertions, translocations, inversions and small-size mutations including point mutations and mutational signatures. The methods of the invention utilize a pool of TArget Capture Sequences (TACS) to enrich for sequences of interest in a mixed sample containing both maternal and fetal DNA, followed by massive parallel sequencing and statistical analysis of the enriched population to thereby detect the risk of a genetic abnormality in the fetal DNA. Kits for carrying out the methods of the invention are also provided.

The invention provides a capture primer group and a kit for targeted Salivirus genome sequencing and application of the capture primer group and the kit, and belongs to the technical field of genome sequencing. The targeted capture primer group of the Salivirus virus genome comprises a primer 1 and a primer 2, the nucleotide sequence of the primer 1 is as shown in SEQ ID NO. 1, and the nucleotide sequence of the primer 2 is as shown in SEQ ID NO. 2. According to the invention, a universal reverse transcription oligo-T primer and the Salivirus specific primer group provided by the invention are put into the reverse transcription step at the same time, so that not only is a macro transcription effect achieved, but also a Salivirus sequence in a sample is synchronously identified for specific capture. According to the method, the complete Salivirus genome sequence can be efficiently obtained, and the method has good coverage and sequencing adaptability to the virus genome.

The present invention discloses a method, system and medium for target capture in a video sequence scenario, belonging to the field of single-object tracking in images. The method includes: performing target detection on video sequence images to obtain the position and size of the target, and deleting the targets whose sizes are not within the specified range; initializing a multi-object tracker using the positions of the undeleted targets to perform multi-object tracking and cumulative confidence calculation; for each object tracker, when its tracking hit count reaches the first set value and the cumulative confidence reaches the set confidence threshold, output the target it tracks and the cumulative confidence, and when its consecutive tracking miss count reaches the second set value, delete the object tracker; initializing a single-object tracker using the target with the highest cumulative confidence to perform single-object tracking and tracking confidence calculation and output, and cross-verifying the results output by the two types of object tracking to determine whether to output and the output results. This method realizes stable capture and recognition of the target.

To provide information processing equipment and other devices that reduce the workload of security operations. [Solution] An information processing device according to one embodiment of the present invention includes an image acquisition unit that acquires an image of a security target captured by an imaging device, an instruction unit that inputs a prompt to a generation AI model that includes an instruction to output in text the result of comparing a reference image, which is set in advance as a reference image among the images of the security target, the latest image which is the most recent image in the time series, and past images which are images acquired in the past than the latest image, and a notification unit that generates a notification based on the output of the generation AI model.

The application discloses a kind of cross array ballistic measurement radar beam scheduling and task allocation method, to improve the target acquisition and tracking efficiency of radar system.The present application is based on cross array antenna, and two kinds of beam scheduling modes are designed:TWS mode and TAS mode;TWS mode allows radar to track the target captured in real time while searching for targets, suitable for occasions handling multiple targets;While TAS mode first searches comprehensively, then tracks the target of interest, effectively avoids multiple beam emission to no target area.Radar can identify and capture targets in the shortest time, improve response speed;After target track is generated, radar can track targets in real time, ensure the accuracy of data;Reduce the number of invalid beam emission, improve resource utilization efficiency.System can dynamically adjust beam scheduling strategy according to target, so as to realize fast acquisition and accurate tracking target, optimize resource utilization efficiency.

The invention generally relates to capturing, amplifying, and sequencing nucleic acids. In certain embodiments, copies of the sense and antisense strands of a duplex template nucleic acid are captured using linked capture probes and multiple binding and extension steps to improve specificity over traditional single binding target capture techniques. Methods of seeding sequencing clusters with sense and antisense strands of a target nucleic acid are also disclosed including identifying the strands using sense-specific barcodes and confirming base calls using two sense-specific sequencing reads. Linked adapters may be used to increase adapter ligation selectively or efficiency and yield.

The present invention discloses an anti-interference target acquisition method based on machine vision technology measurement, the steps of which include: S1, original image multi-stage filtering preprocessing; S2, target feature extraction; S3, random forest construction: calling a historical data sample library to obtain an anti-interference target acquisition library, randomly selecting M data points from it, creating a decision tree for the selected data points, each decision tree will produce a result, comprehensively analyze the results and output a majority vote or average; S4, through random forest, obtain a target feature reference set; S5, select the first N frames of the on-site detection picture as a training set, obtain appropriate counting weights through continuous training, and then start from the N+1th frame image, perform voting analysis by determining the weight and using the Adaboost algorithm to determine the final target. The present invention can effectively solve the interference problem in the target capture process under complex scenes, and can effectively improve the accuracy and robustness of target detection.

A projection system and a projection method are provided. The projection system includes a projection module, an image capture apparatus, an input apparatus, and a processor. The processor is configured to: receive an input signal from the input apparatus and generate an input command accordingly; drive the image capture apparatus to obtain a target captured image, where the target captured image includes a target image corresponding to a target object; transmit the target captured image and the input command to a cloud server; receive an analysis result generated by an artificial intelligence computing module according to the input command and the target image from the cloud server; determine display information according to the analysis result, and calculate a display position of the display information within a projection range; and drive the projection module to project an information image with display information.