1810 results about "Multi targeting" patented technology

The invention discloses an intelligent multi-target active tracking monitoring method and system. The method comprises a plurality of steps such as a multi-target active tracking step, an active focusing step, a target switching step, a relay tracking step, an access warning step and a complicated environment preprocessing step. A panoramic picture in a monitor region is acquired by virtue of a first camera, a tracked target is locked, real-time coordinate information of the tracked target in the panoramic picture as well as a first angle control value of a cradle head of a second camera corresponding to the coordinate information and a first focusing value of the second camera are calculated, when a close-up image of one tracked target needs to be acquired, the second camera positions the tracked target according to the angle control value and the first focusing value corresponding to the selected tracked target so as to acquire the close-up image, so that the second camera is always focused on the tracked target to continuously track and photograph the tracked target in real time. By adopting the intelligent multi-target active tracking monitoring method and system, the cradle head control problem and focusing problem in the video monitoring and multi-target tracking can be effectively solved.

The invention provides an automotive anti-collision radar multi-target detecting method and system. The method comprises the steps of emitting two types of triangular waves with different modulation periods in an alternating mode and acquiring echo data, carrying out windowing processing, carrying out distance dimension FFT and speed dimension FFT, carrying out modulo processing to obtain frequency spectrum of two types of echo waves, carrying out target paring to obtain a spectral line of the same target of the two echo waves, calculating distance and speed of each target, and judging target distance and speed obtained by the two echo waves through a tolerance function to obtain a final target. A radio frequency emitting and receiving part of the system comprises a radar sensor and an intermediate frequency processing module, and a data processing part comprises a modulus, a modulus converting module and a central control processing module FPGA. The FPGA comprises a modulation signal producing sub-module, an echo wave signal acquisition sub-module, an algorithm sub-module and a control sub-module. The modulation waves and corresponding algorithms effectively remove false targets and improve accuracy that multiple moving targets are detected under strong noise. A hardware system is simplified in structure and easy to achieve.

The invention provides a route planning method based on a multi-target glowworm swarm algorithm and belongs to the technical field of route planning. The method includes: modeling a route planning problem, initializing the multi-target glowworm swarm algorithm, updating the glowworm position, determining a non-inferior solution set, updating an external archived file, judging whether a preset maximum iteration number is achieved or not, and determining a Pareto optimal route. The basic glowworm swarm algorithm is improved based on the Pareto dominant conception, and global searching and parallel computing capacities of the glowworm swarm algorithm are well used. Multiple route performance indexes are considered simultaneously during planning, a group of Pareto optimal solution sets can be obtained by planning once, and high flexibility is achieved. Further, the route panning method is different from traditional route planning methods for a single target and route planning methods of using a weighing method to convert multiple targets into the single target, and can better meet practical requirements of route planning.

The invention provides a named entity identification method capable of combining an attention mechanism and multi-target cooperative training. The method comprises the following steps that: (1) carrying out a preprocessing operation on training data, and through character hierarchy mapping, obtaining the character vector representation of a sentence; (2) inputting the character vector representation obtained in (1) into a bidirectional LSTM (Long Short Term Memory) network, and obtaining the character vector representation of each word; (3) through word hierarchy mapping, obtaining the word vector representation of each sentence; (4) through the attention mechanism, splitting the word vector representation obtained in (3) with the character vector representation obtained in (1), and transmitting into the bidirectional LSTM network to obtain the semantic characteristic vector of the sentence; and (5) aiming at the semantic characteristic vector obtained in (4), carrying out entity annotation on each word by a conditional random field, and decoding to obtain an entity tag.

The invention discloses a traffic scene multi-target classification method, to be specific, discloses a traffic scene classification method based on a multi-scale convolution neural network. The traffic scene classification method is characterized in that recessive characteristics based on the multi-scale convolution neural network are extracted; and an optimal covering segmentation tree is acquired. During the realizing of the traffic scene classification, the multi-scale convolution neural network is adopted, and the excellent recessive characteristics having the invariance property are effectively extracted from an original image in different scales, and by comparing with the single-scale convolution neural network, the acquisition of the abundant and effective characteristic information of the image is realized. The effective information extracted by the convolution neural network is combined with the original segmentation tree of the image to form an optimal purity price tree, and the covering having the optimal purity is carried out, and therefore a clearer target contour is acquired, and the classification accuracy is increased. The RGB-D is used as the convolution neural network input, and by comparing with the conventional RGB convolution neural network input, the training characteristic is additionally provided with the depth information, and the classification of the input image is more accurate.

The invention relates to an individuation catering recommendation method and system based on multiple targets. The individuation catering recommendation method mainly includes the steps that essential information and behavior information (including diet records and browsing behaviors on webpages) of a user are collected and recorded into a database; a user model is built according to user information in the database, and individuation nutrition recipes are recommended; nutrition balanced diet recipes are generated through multi-target optimization catering models, the similarity between the recipes based on nutrient elements is calculated through a collaborative filtering recommendation model and an equivalence interchange model, various interchange recipe lists are generated from recipe bases, and the dietary structure is enriched; the nutrition balance of the recipes is detected, improvement measures are provided by comparing practical contents and recommended nutrient intakes of various nutrients in the generated recipes, and the designed recipes trend to be reasonable.

An information processing system for branch target prediction is disclosed. The information processing system includes a memory for storing entries, wherein each entry includes a plurality of target addresses representing a history of target addresses for a multi-target branch and logic for reading the memory and identifying a repeated pattern in a plurality of target addresses for a multi-target branch. The information processing system further includes logic for predicting a next target address for the multi-target branch based on the repeated pattern that was identified.

The invention discloses a method for identifying abnormal motion in a video. The method comprises the steps that an area of which the pixels have certain change is extracted from a video sequence through a foreground extraction module; a movement target detected by the foreground extraction module is judged through a pedestrian judgment module, and a rectangular box judged to be a pedestrian is kept and transmitted to a target tracking module; each identified target in the scene is continuously tracked in a multi-target way through the target tracking module; the posture of each target tracked by the target tracking module within each frame of time is judged through an abnormal behavior trigger module and behavior analysis is performed on the abnormal behavior; and the abnormal behavior is processed and computed into a video segment through a behavior identification module, and the abnormal behavior is alarmed and the behavior type of the abnormal behavior is identified. The invention also discloses a system for identifying the abnormal motion in the video. The problem that abnormal behavior (falling down) detection of massive video data cannot be performed in real time can be solved and the detection accuracy can be greatly enhanced.

The invention discloses an electric vehicle travel planning method based on multi-target optimization, generally comprising the following steps: (1) a travel planning problem model is established; (2) drivers provide travel information; and (3) an optimal scheme is solved based on a timed multi-target ant colony optimization algorithm. The problem model comprises a road network model, a vehicle model, and travel target and travel constraint definition. Travel information includes: not providing any information, providing constraint information, and providing optimization goal and constraint information. The ant colony optimization algorithm includes the steps of pheromone initialization, route transfer probability calculation, travel scheme search, air conditioner use determining, travel scheme ranking, pheromone updating, and loop optimization. A dynamic stochastic road network model is used to describe the traffic environment and plan the travel of electric vehicles, and target characteristics corresponding to different travel schemes can be reflected. The ant colony optimization algorithm ensures that a multi-target and multi-constraint optimized electric vehicle travel scheme is generated as the number of iterations increases.

The invention belongs to the field of path planning technology and specifically relates to a three-dimensional multi-UAV coordinated path planning method based on SAS. The method comprises the following steps: carrying out modeling on a path planning environment; initializing multi-target SAS calculation parameters consisting of the length of a minimum path section, a maximum turning angle, a maximum angle of climb/glide, a minimum safe distance of UAVs and a minimum flight altitude of UAVs; initializing the positions of the UAVs, wherein each UAV represents a path; updating the positions of the UAVs; expanding a current node; determining whether a path section collides with other path sections; updating a node table of the path section; executing a step (8) if minimum path cost set in the step (2) is met, and otherwise, executing the step (3); determining a cooperated planned optimal path so as to complete path planning. The method overcomes multi-objective optimization problems, has versatility, provides a reasonable optimal solution for a decision maker and better accords with practical needs.

The present invention discloses a multi-year regulating storage reservoir optimal scheduling method considering runoff uncertainty. The method comprises the following steps of: (1) analyzing historical reservoir runoff data; (2) establishing a runoff random simulation model, and generating a reservoir simulation annual runoff sequence; (3) simulating annual inflow runoff sequence classification and performing intra-annual distribution to obtain a three-year inflow runoff process sequence on a month scale; (4) establishing a multi-year regulating storage reservoir optimal scheduling model that considers multiple water demands; and (5) reducing dimension by using an optimization algorithm and optimizing a resolution model so as to obtain an optimal discharging and water supply decision. According to the method provided by the present invention, the scheduling with maximized comprehensive benefits of long-term multi-target operation can be performed on a multi-year regulating storage reservoir by considering runoff uncertainty, so that the method is suitable for promotion in long-term optimal scheduling of the multi-year regulating storage reservoir in China.

The invention discloses a high-rigidity and light-weight design method considering uncertainty of a slide block mechanism of a press machine. The high-rigidity and light-weight design method comprises the steps of: establishing a high-rigidity and light-weight design model considering the uncertainty of the slide block mechanism of the press machine; adopting an optimized Latin hypercube sampling method for a test design, obtaining a target function response value corresponding to each sample point through collaborative simulation and building a Kriging agent model; converting an uncertain target function into a certain target function based on an order relation of interval numbers; and calculating a target function interval by utilizing an internal structure analysis way, finding a Pareto optimal solution collection of a converted certain optimal problem by utilizing a multi-target genetic algorithm, if the precision requirement is not satisfied, performing important sampling at the place where the extreme value of the target function interval locates, and updating the target function sample collection and the agent model to perform iterative optimization. The high-rigidity and light-weight design method considering uncertainty of the slide block mechanism of the press machine, disclosed by the invention, has the advantages of establishing the uncertain optimization model of the slide block mechanism according with the practical project for solving and really realizing the light-weight and high-rigidity design of the slide block mechanism.

The invention belongs to the field of intelligent traffic video image monitoring and video image analysis, and in particular relates to a method and a system for automatically identifying an urban traffic accident. The method for automatically identifying the traffic accident comprises the following steps of: acquiring an urban road video image sequence; performing foreground vehicle separation based on a mixed Gaussian background model; performing a multi-target vehicle tracing algorithm based on a Camshift algorithm and a kalman filtering combination; extracting traffic accident determiningparameters such as speed variation, horizontal position variation, vertical position variation, moving direction variation and the like; and proposing a multi-featured weighted fusion automatic accident identification algorithm. Traffic accident information is transmitted to a traffic control center in time by a transmission unit and a display unit, so that the traffic accident can be quickly treated, an effective and flexible road traffic monitoring means with high cost performance is provided for traffic management, and new thought is provided for the development of a high-efficiency intelligent video traffic accident system.

Provided is a multi-target route planning device and method considering searching robot target node timeliness. A multi-target genetic algorithm is used to achieve route planning considering two optimization objectives including route consumption and the node timeliness through two stages of decoupling which comprise the route planning and route generating, and therefore improvement of searching performance of a mobile robot which is supervised by an operator is benefited. Particularly, when the problem of the node timeliness needs to be considered in the process of route searching, a good searching manifestation can be obtained.

An object oriented framework for source code processing provides an infrastructure that embodies the steps necessary to process source code and a mechanism to extend the framework to fit a particular scenario for processing source code. Certain core functions are provided by the framework, which interact with extensible functions provided by the framework user. The architecture of the framework allows a developer to select the native processing tool of their choice, and provides an infrastructure for using that processing tool with other platforms (e.g., targets, etc.) The extensible functions allow the processing tools of a new platform to be easily incorporated into the framework. The framework thus allows a programming environment with development on multiple platforms that targets multiple platforms, which may be easily customized to include new platforms and new processing tools.

The invention discloses a micro-grid capacity programming method meeting a coupling type electric cold and heat demand, comprising: establishing a micro-grid capacity energy supply device capacity programming model meeting a coupling type electric cold and heat demand, and a micro-grid capacity energy supply device optimum capacity programming calculation method based on the multi-target particle swarm technology. According to the mode, the method solves the problem that a current programmed micro-grid needs to simultaneously provide power supply and cool supply (or heat supply) for users, and the problem of capacity optimum allocation among different energy supply devices when a coupling relation exists between electric cold supply and electric heat supply. The model realizes multi-target optimization under the condition of considering engineering, economy and security constraints of micro-grid energy supply devices.

The invention discloses a key target identification method for an automobile cruising system. Based on the fuzzy control theory, a multi-target automobile lane changing fuzzy logic controller is designed, a dangerous lane changing automobile is prejudged, and effective identification of a straight lane key target is determined. The tilting motion influence of automobile bend running is considered, an extension Kalman filtering on-line real-time estimation of an automobile mass center side slip angle and a road curvature is achieved, the bend key target identification method is provided, a key target judgment basis is given, and effective identification of the bend key target is achieved.

The invention discloses an automatic driving control system and method. The control system comprises a curve synthesis unit, a feedback unit, a controller and an optimization unit, wherein the curve synthesis unit is used for calculating an object speed curve according to a condition inputted by an external system; the feedback unit is used for acquiring state information during the operation process of a train; the controller is used for making a train control instruction according to the object speed curve and the sate information fed back by the feedback unit; and the optimization unit is used for optimizing the control instruction outputted by the controller by use of an instruction optimization strategy and a relation matrix combined and constructed by conditions to obtain an optimized control instruction. According to the invention, through increasing optimization units, multi-target data is obtained, i.e., multiple parameters are obtained, and the object speed curve obtained through calculating by the curve synthesis unit is formed, the control instruction is obtained through combination with the state information which is fed back, and the control instruction is optimized, so that in a specific scene, self-adaptive automatic driving control is realized, the control of a vehicle-mounted controller is facilitated, and the control effects are improved.

The present invention relates to a target tracking cooperative control system and method based on multiple unmanned surface vehicles. The system is formed by connecting a shore-based global location host and a single unmanned surface vehicle control system through a wireless communication module. The method comprises the operation steps of: 1) a formation generation process: employing an auction algorithm to find a multi-target distribution scheme of the multiple unmanned surface vehicles with the maximum income of an unmanned surface vehicle group; 2) motion of the unmanned surface vehicles to perform geometric path planning from any initial state to a target point; and 3) prediction of a target motion track through adoption of a prediction model based on a particle swarm to replace communication abnormal data and perform formation track tracking. The method reduces the calculation amount of the multiple auction processes, achieves the real-time demands of task distribution of the unmanned surface vehicles, employs the path planning method based on the geometric method and the track tracking method based on the neural network to meet the timeliness and the accuracy requirements ofsingle-vehicle track tracking control, employs the motion track predicted by employing the particle swarm optimization to perform compensation, improves the tracking capacity of the unmanned surfacevehicles in the limitation of the communication condition and allows the formation tracking to have high reliability and stability.

The invention discloses an unmanned surface vessel path following guidance method considering hybrid multi-target obstacle avoidance. The unmanned surface vessel path following guidance method considering hybrid multi-target obstacle avoidance includes the steps: dividing the guidance process into a path following mode and an obstacle avoidance control guidance mode to improve a DVS guidance algorithm as the basic framework; dynamically programming a smooth reference path formed by lines and curves by means of a GVS, wherein the path following mode and the obstacle avoidance control guidance mode are respectively corresponding to different guide vectors of the DVS; and for obstacle avoidance guidance of a plurality of or hybrid obstacles, determining the current obstacle avoidance target according to the priority sequence and obstacle avoidance control conditions, and starting the obstacle avoidance control mode, wherein the transition function guarantees the smoothness of the DVS guide vectors for switching among different modes. The unmanned surface vessel path following guidance method considering hybrid multi-target obstacle avoidance has applicability for various control strategies and is convenient for combination with the current advanced control algorithm, wherein the control algorithm is used for guaranteeing convergence of DVS for a real ship and guaranteeing effectiveness of the guidance strategy.

The invention relates to a new technology for depositing a CrAlSiN gradient coating with high-temperature oxidation corrosion resistance and excellent mechanical properties on a supercritical water-cooled reactor fuel clad surface in a nuclear fission reactor. The CrAlSiN gradient coating is deposited on the surface of a base material by adopting a multi-target reaction magnetron sputtering method, and the technology comprises the following four continuous stages of: 1, preparing a Cr gradient coating; 2, preparing a CrAl gradient coating; 3, preparing a CrAlN gradient coating; and preparing the CrAlSiN gradient coating, and annealing. By reasonably designing the gradient variable microstructure of each element, the anti-oxidation temperature of the coating reaches 950 DEG C, the hardnessof the coating reaches over 37GPa, and the adhesive force of the coating adhered to the base material reaches 34N; the coating also has excellent thermal shock resistance, abrasion resistance and thelike; and the service property of the supercritical water-cooled reactor fuel clad part can be greatly improved, and the service life of the clad part can be greatly prolonged.

The invention discloses a wind, light and water-containing multi-source complementary micro-grid hybrid energy storage capacity optimal proportion method. According to the method, an annual output power curve of wind power generation, photovoltaic power generation and hydroelectric generation is simulated according to the distribution condition of natural resources such as wind, light and water, an annual load curve of a micro-grid is combined, system cost and power fluctuation are used as target functions, accumulator capacity and super-capacitor capacity are used as optimization variables, and meanwhile constraint conditions such as power balance constraint, maximum instantaneous power constraint, power supply reliability constraint, super-capacitor charge and discharge current and voltage constraint and accumulator SOC (System On Chip) constraint are determined to establish a wind, light and water-containing micro-grid hybrid energy storage optimization configuration model; optimized solution of the target functions is performed by using a fuzzy decision-containing multi-target planning GA-PSO (Genetic Algorithm-Particle Swarm Optimization) algorithm to obtain the optimal proportion of the hybrid energy storage capacity. Compared with the conventional GA algorithm and PSO algorithm, the method has the advantages that the convergence rate is higher and the problem of mutual conflict of the target functions in the multi-target optimization algorithm is avoided better.

The invention relates to a resource management method suitable for a multilayer satellite system. The resource management method adopts the following design principle that high altitude satellites are used as reserved resources and are specially used for distributing resources for users with high priorities and high QOS (Quality of Service) requirements and distributing the resources for the user as required in a real-time manner; and meanwhile, low earth orbit satellites are combined and are used as ground network supplements, the characteristics of high-speed motion and the like of the low earth orbit satellites are considered and a method for dynamically distributing the resources for users with low priorities by using efficiency, cost and user publicity as evaluation indexes is designed. According to the design, the characteristics of the high altitude satellites and the low earth orbit satellites are combined; the priorities of the users are considered; the resources are distributed for the users with high priorities as required; a multi-target evaluation function is designed for the users with low priorities; and the multi-target evaluation function is nondimensionalized, so that in the resource management process, the evaluation indexes of other targets are not ignored due to an excessive function value of a certain index, and thus, a multi-target optimization method which has the advantages of short time delay, high bandwidth utilization rate and balanced resource distribution is integrally implemented.

The invention discloses a multi-target detection model construction method based on a convolutional neural network to solve the technical problem that the existing detection model cannot distinguish multiple targets and is difficult to identify small targets. The method comprises the following steps: 1: constructing a Caffe deep learning framework, wherein the configuration of the detection modelis completed by using a Faster R-CNN algorithm, and a ZF network is introduced for feature extraction; 2, designing an ADPN network to accurately generate a multi-target area in real time; 3, designing and optimizing the loss function of the ADPN; 4, training the ADPN; 5, designing a DALN sub-network for detecting multi-target categories and locations; 6, designing and optimizing the loss functionof the DALN; 7, training the DALN; and 8, jointly training the ADPN and the DALN to obtain a detection model. The method can identify the targets of multiple categories, improve the recognition ability for the small targets, and is fast in calculation speed and high in precision.