81 results about "Heuristic (computer science)" patented technology

The invention discloses a real-time gesture interaction method based on computer vision, which comprises the following steps: acquiring a digital image from an image input device; detecting hands by a statistical learning method; initializing a hand contour tracker according to the detecting result, and calculating a skin color model of a specific hand; tracking the hand contour by a combined method of a conditional probability density transmission algorithm and a heuristic search technology; analyzing the moving track of the hands by a Hidden Markov Model to obtain the gesture identifying result; and applying the gesture analyzing result to the interaction of various application programs. The real-time gesture interaction method of the invention expands the interactive mode of the traditional mouse and keyboard, realizes automatic hand detection, tracking and gesture identification by the computer vision and image processing techniques, has real-time performance, robustness and easy realization and operation, and can enable computer users to interact with the computer more naturally, visually and intelligently by hand gestures.

The invention relates to a method for automatic fault diagnosis of a satellite and a diagnostic system thereof. The method comprises the following steps: dividing the satellite fault into a component level fault, a subsystem level fault and a system level fault; adopting a fault detection method based on a fuzzy set to fuzzify a precise sensor sampling value; defining different alarm weight values for fuzzy subset based on the membership grade belonging to each fuzzy subset; and determining whether to alarm or not by calculating the relation between the summation of the alarm weight values in a time window and an alarm threshold value. The method can effectively overcome the defects of a common fault detection method and effectively reduce the false alarm rate and the rate of missing report; and when certain fault is caused by a plurality of sub-faults, the method performs heuristic search according to the reliability of a system corresponding to each sub-fault and searches a sub-fault node corresponding to a system with low reliability preferentially. Besides, in order to adapt to the synergistic diagnosis between distributed satellites in the future, a synergistic interface is designed to provide diagnosis services for other satellites, thus the aim of the invention is achieved.

An improvement to a predictive video encoding method or apparatus that includes selecting a mode class within which to choose an encoding algorithm to encode individual blocks of video information. Rather than implementing each algorithm in each mode class to ascertain an acceptable compression, the improvement eliminates searching through a class of encoding modes based simply on heuristics. The method comprises obtaining statistical information related to previous blocks of encoded video information, determining a mode class within which to chose a particular encoding algorithm based on the statistical information (e.g., heuristics) gathered, choosing an algorithm within the selected mode class using conventional techniques, and encoding the video information according to the chosen algorithm. Statistical information may include quantization parameters, prior encoding decisions, intensity or frequency values, or Hadamard transform coefficients of previously encoded macroblocks. Encoding complexity in the coder is reduced since one class of encoding modes is eliminated.

The invention discloses a mobile robot navigation method based on a binocular camera and a two-dimensional laser radar. The mobile robot navigation method includes the steps that a two-dimensional grid map model is built; a mobile-robot pose model is built; a laser radar data model is built; a joint calibration model of the binocular camera and the laser radar is established; route planning and obstacle avoidance navigation are carried out with the Dijkstra algorithm and the A-Star algorithm. According to the mobile robot navigation method, the relative position and the relative orientation ofthe binocular camera and the two-dimensional laser radar can be determined with the triangle joint calibration method; in the known-environment two-dimensional grid map environment, the binocular camera can be used for detecting barriers higher than and lower than the two-dimensional laser radar plane, the barriers can be mapped into an environment map, through the shortest path algorithm and heuristic search optimization, a navigation path line capable of avoiding the barriers is obtained, and a mobile robot can reach the destination along a path capable of avoiding the barriers in the movement process.

Methods and systems are described for computing shortest paths among a set of locations. A small set of landmarks is chosen and the distance between each location and each landmark is computed and stored. Given source and destination locations, the landmark distances are used to compute lower-bound estimates of distances from locations to the destination. The estimates are then used with a heuristic search to find the shortest path from source to destination.

The video moving target self-adaptive tracking method based on multiple characteristic information fusion can be extensively used in computer vision, image processing and mode identification field. Said method includes the following steps: firstly, extracting characteristic information of target, then utilizing fuzzy logic method to make self-adaptive fusion of various characteristic informations, using fused information to describe observation information of target, finally adopting genetic algorithm with heuristic search function to find out candidate target most similar to target template observation information in current image target candidate zone.

The invention relates to a real-time information-based metro passenger service system. Mobile terminal equipment automatically connects wireless sensors in a metro station and in metro compartments so as to perform data interaction with the servers to acquire relevant information in the metro station in real time, and the relevant information is used for passengers to view in real time; moreover, a heuristic search strategy based on the estimated arrival time of a metro, running time between stations, the current passenger flow information of the metro station and passenger flow information in each metro compartment is proposed, and therefore the optimal proposed ride scheme is obtained, and meanwhile, a time-optimal and distance-optimal scheme is obtained for the passengers to select; after the scheme is determined, according to the position conditions of the current passengers, a prompt of the next route is provided in real time; the passenger flow conditions of the compartments are determined according to the information of the wireless servers of the metro which is about to arrive, and the best waiting position proposal can be given by the system; after the passenger ride, prompt contents, including the time that the metro is about to arrive the next station, the get-off station and the get-off time of the passengers are given in real time, the exit scheme of the metro station and corresponding map information, can be given in real time.

The invention discloses a method for fault component-based wide area backup protection, which belongs to the field of relay protection for electric power systems. The method comprises the following steps of: when a power network is in normal operation, dividing protection associated domains according to a topology structure and the PMU configuration of a network; updating the protection associated domains in real time by adopting a heuristic search technique according to switch transposition information; when a fault happens, extracting fault stable-state components of current and voltage, and defining fault associated domains according to the fault stable-state component of differential current in each associated domain; and calculating fault associated factors in the associated domain in real time to determine fault branches. The method can accurately position the faults under a condition with PMU limited measuring points, is not influenced by the system operation mode, avoids the condition that the conventional backup protection can have mismatching or insufficient sensitivity very easily, is not influenced by load current, and has very high reliability.

A machine-learning based artificial intelligence device for finding an estimate of patent quality, such as patent lifetime or term is disclosed. Such a device may receive a first set of patent data and generate a list of binary classifiers. A candidate set of binary classifiers may be selected and using a heuristic search, for example an artificial neural network (ANN), a genetic algorithm, a final set of binary classifiers is found by maximizing iteratively a yield according to a cost function, such an area under a curve (AUC) of a receiver operating characteristic (ROC). The device may then receive patent information for a target patent and report an estimate of patent quality according to the final set of binary classifiers.

The invention discloses a heuristic route segment path-finding method for ship route design, comprising heuristic search, a heuristic shortest route segment path-finding algorithm analysis method, an A* algorithm heuristic principle and an improved shortest path algorithm, wherein the heuristic route segment path-finding algorithm puts forwards a method establishing a navigable path tree to reduce meaningless path searching time by being based on a point to point navigable line algorithm and combining with the A* algorithm heuristic principle. By analyzing the particularity of the route segment path-finding algorithm based on a vector diagram in the ship route design, the route segment path-finding algorithm is improved according to an analysis result, the accuracy of a generated route segment path is enhanced and the path-finding efficiency is further increased. Lands, islands and other barriers positioned on a map of the world are expressed by polygons in the vector diagram, and route segment path-finding is to finally find an optimal route segment from communicated paths which are generated between the two outer points of the polygons without passing through the polygons.

Methods of recovering data in a received signal sent in a communications media are disclosed. Composite channel impulse responses are first estimated. Channel-tap locations are then assigned to suppress the interference noises by sequential search schemes or heuristic search schemes based on estimated composite channel impulse responses. A sequential search scheme optimizes a predetermined design criterion in a sequential manner. Also described are recursive evaluations of the design criterion and the inverses of the noise covariance matrices based on the composite channel impulse response during a sequential search. A heuristic search scheme selects channel-tap locations based on a set of pre-selected channel-tap locations. The set of pre-selected channel-tap locations is determined according to the estimated composite channel impulse response. A method of estimating energy levels of known interference sources is also described.

The method for Hammerstein modeling of a steam generator plant includes modeling of the linear dynamic part of a Hammerstein model with a state-space model, and modeling the nonlinear part of the Hammerstein model with a radial basis function neural network (RBFNN). Particle swarm optimization (PSO), typically known for its heuristic search capabilities, is used for estimating the parameters of the RBFNN. Parameters of the linear part are estimated using a numerical algorithm for subspace state-space system identification (N4SID).

The invention discloses a wind power unit commitment contained modeling method considering predication error timing sequence distribution. A predication error segmentation fitting method under a timing sequence based on error characteristic analysis is proposed; fitting is performed based on t location-scale distribution to reduce a heavy-tail effect and improve the fitting precision, and the fitting method can be matched with a UC decision in timing sequence; next, a dual-quantile type UC decision model capable of taking conventional cost, extra standby cost and risk cost into considerations is established; the selection of confidence levels can be balanced through the restrictive relations among different costs; the standby classification can be guided by the division of the different confidence levels; the error timing sequence segmentation distribution is adapted by time varying confidence levels, so that the model is more economical, targeted and applicable; and finally, an improved hybrid particle swarm algorithm with a heuristic searching principle is adopted to solve a multi-variable mixed integer programming model of a text, and the effectiveness of the method is verified by example results.

The invention discloses a method for determining automatically mosaic line of orthophoto and performing automatic seamless mosaic of orthophoto, including selecting the optimum route to evade obstacle area in a difference image using the positive feedback of ant algorithm and elicitation type search characteristic of swarm intelligence; obtaining the mosaic lines of two orthophotoes to be spliced, performing mosaic fusion to the orthophoto pairs using the mosaic lines, thereby implementing the automatic seamless mosaic of orthophoto. The method which determines automatically the position of mosaic lines in the orthophoto mosaic process is capable of improving greatly the production efficiency of orthophoto map in mapping industry by replacing the prior method for determining or modifying mosaic lines manually, in which the mosaic lines evades automatically ground feature higher than the ground, such as houses, crown of trees, and the like, and evades the areas with large contrast of imaging colors in adjacent orthophoto maps.

A system and methods for multi-language abstract digital simulation model generation and execution, comprising a meta-model structuring and creation system, meta-model mapping table, remote server, simulation execution process, computer domain specific language, and methods for user-creation and editing of meta-models, simulation models, and parametrization of simulation environments, actors, objects, and events in real-time using heuristic searching.

The invention provides a method and a system for sensing multiple-dimension electric utilization activities. The method comprises the following steps of step 1, predefining statistical features of electric utilization equipment under a plurality of dimensions, obtaining a feature value of each statistical feature, and establishing a corresponding statistical model; step 2, installing a collecting unit at an entrance of an electric network, obtaining the current and voltage information of all electric utilization equipment in a real-time way, and comparing and training the feature weight of each statistical feature according to the current and voltage information and input data of an user by a calculating unit; and step 3, carrying out the enlightening type searching and calculation according to the feature values, the statistical models and the feature weights by the calculating unit, so as to obtain the real-time electric utilization activity of each electric utilization equipment, and sending the real-time electric utilization activities to a display unit. The method and the system have the advantages that multiple load features are combined with a series of statistical calculating methods to directly obtain the information of the electric utilization activities of all household equipment, the resolution capability is high, the judgment precision is high, and the electric utilization activities of the similar electric utilization equipment can be effectively distinguished.

The invention provides a freight rate searching system which comprises a travel experience information module, a freight rate index module, a freight rate route selection module, freight rate splitting module and a freight rate combination and verification module, and the travel experience information module is used for generating a travel experience information database; the freight rate index module is used for establishing a freight rate index so as to search for the freight rate rapidly; the freight rate route selection module is used for generating a potential low-price route, and the lowest price in a freight rate set of the corresponding precision is extracted according to travel information input by a traveler; the freight rate splitting module is used for generating price units and freight rate journeys; the freight rate combination and verification module is used for generating the final optimal freight rate combination. By means of the freight rate searching system, the purpose of flight search guided by price is achieved, a heuristic search mode is adopted, price and comfort of a potential journey are estimated through travel experience data, unreasonable results are eliminated, and the final result meeting the requirements of travelers is screened out.

The present invention provides a door-to-door travel path scheme personalized recommendation method based on heuristic search. The method comprises the steps of (1) according to a real-time traffic network diagram, initializing a candidate-path set, (2) carrying out iterative updating of the candidate-path set until K candidate paths are obtained, (3) sorting the K candidate paths in the candidate-path set, and (4) outputting the sorted K candidate paths as recommendation paths. According to the method, K optimal shortest paths are searched in a short period of time to carry out personalized recommendation, the preference information of passenger is mined in a relatively short period of time, a best travelling path is recommended, the overall effect of the personalized recommendation is improved, and the traveling of the passengers is facilitated.

The invention mainly relates to the technical field of quick satellite earth observation, in particular to a method for realizing hybrid and active scheduling on quick satellites. The method comprises the following steps: S1, generating an initial flexible scheduling plan: alternately performing heuristic search and constraint propagation until processing on all tasks is completed, wherein the resource allocation problem is solved by the heuristic search, and the activity start time range is reduced by the constraint propagation; S2, performing hybrid and active adjustment on the scheduling plan: performing scheduler underlayer processing by utilizing the minimum disturbance updating method of the constraint propagation when the task observation start time, the activity duration and the activity observation posture are adjusted; performing adjustment on a task observation time window and insertion of tasks, which are not arranged, by utilizing the heuristic search method for inserted tasks. The method for realizing the hybrid and active scheduling on the quick satellites, disclosed by the invention, is simple in operation, high in flexibility and easy to realize, and can be applied under the complicated background that the multiple quick satellites are scheduled.

A method for combining a random set of video features non-linearly to evaluate perceptual quality of video sequences includes (a) receiving a video sequence for image quality evaluation; (b) providing an objective metric image quality controller comprising a random set of metrics ranging from M1 to Mn without dependency information for each one metric; (c) applying each one metric individually to the video sequence to provide an individual objective scoring value of the video sequence ranging from x1 to xn; (d) determining a plurality of sets of weights (w1 to wn) which correlate to predetermined subjective evaluations of image quality for a predetermined plurality of video sequences (n), each one set of weights being assigned a range having an incremental value equal to the range divided by a number of combinations for each one set of weights; (e) weighting each individual objective scoring value x1 to xn provided by each one metric of the random set of metrics in step (c); (f) combining metrics of the weighted individual objective scoring value of the random set of metrics into a single objective evaluation F, wherein each weighted individual scoring value from step (e) is multiplied by each individual objective scoring value x1 to xn from step (c); (g) calculating a correlation factor R to provide a correlation value for the objective evaluation F and the plurality of video sequences (n). Steps (e), (f) and (g) are repeated to provide a plurality of correlation factors which are ranked. A heuristic search uses a genetic algorithm to find the best set of weights to provide objective scores closest to predetermined subjective evaluations. A system provides the hardware and modules that perform the non-linear combination of metrics to provide enhanced perceptual image information.

The invention provides a hybrid task scheduling method and model for a railway locomotive operation control system. The scheduling method takes one frame period as a basic scheduling unit; the scheduling method includes the following steps that preprocessing is conducted; according to a periodic real-time task, based on the table-driven secondary priority rule, scheduling sequence sorting is conducted; according to a non-periodic real-time task application, based on a heuristic searching strategy and a fuzzy control method, arrangement is conducted; recycling of time slices and collecting work of scheduling results and feedback data are conducted, and the remaining execution time is determined; if the periodic time is not used up, a non-real time task is executed. The hybrid task scheduling method and model can greatly reduce the cost of a system, can more flexibly handle various change conditions in the system executing process, and can reduce the influence of the system uncertainty on scheduling.

The invention discloses a self-heuristic type strategy-based natural scene character detection method and system. According to the method and system, a two-level cascaded filtering mechanism is established to judge character areas and background areas; provided that each character area has a dark-background light-character mode or a light-background dark-character mode in a gray scale natural scene, the contrast types of the character areas are judged, so that the contrast of the character areas in the gray scale natural scene is unified to be the dark-background light-character mode; and results which are judged as characters after two-level cascaded filtering processing is performed are adopted as seed characters; self-heuristic type search strategies are constructed at regions near the seed characters for each seed character; a trained deep neural network is adopted to detect undetected charters through sliding a detection window so as to improve the recall rate of character detection; and adjacent characters are separated from each other with words adopted as units. The self-heuristic type strategy-based natural scene character detection method and system of the invention is of great practical value in the semantic automatic understanding of characters in the natural scene.

The invention relates to a heuristic search method with exact solutions to shortest path problems based on place learning effect, in particular to a robot path planning method. In finding a shortest path, a robot interacts with an environment to acquire experiences through reinforcement learning, which means the robot may acquire a same path at very low cost; therefore, in such case, the shortest path may be different from a shortest path with no learning effect. In order to solve the problem, after the robot acquires external environmental information and determines a learning effect function according to priori knowledge, pruning operation and filtering operation satisfactory to this problem are designed through heuristic information to eliminate in advance part of the paths which are determined not likely to present on the shortest path, the search process is accelerated, and the robot may find an exact global shortest path within reasonable effective time to guide its travelling.

The invention discloses a method for searching for a heuristic shortest path based on direction optimization. The method includes the steps that directivity selection is performed on expandable nodes of current nodes, and a range in searching for nodes is reduced; path estimated values of nodes meeting the requirements are compared according to a heuristic searching function, nodes with low option values are preferentially selected to be expanded, and father node information of the nodes are recorded. With the depth-first method, path nodes are searched for in a step-and-step deepening mode till all the expandable nodes are traversed. The method for searching for the heuristic shortest path based on direction optimization has the advantages that the direction optimization principle is used, so that the traversing number of the nodes is greatly reduced, and the searching speed is improved; meanwhile, the heuristic function is introduced so as to reduce blindness in expanding the nodes with the depth-first method, the optimal node is preferentially considered to be expanded so that the shortest path can be found in early-stage searching, and multiple alternative paths can be provided.

The present invention relates to a technical innovation process management method based on knowledge grid, belonging to the field of knowledge management (KM). Said method includes the following steps: firstly, analyzing enterprise information and extracting knowledge element, knowledge chain and knowledge point from said enterprise information, storing said knowledge element, knowledge chain and knowledge point into a knowledge base; utilizing grid technique, using knowledge chain as 'net wire', using knowledge element main body as 'net point' and making them be connected together, using knowledge point as 'entrance' of said grid; then combining the knowledge element, knowledge chain and knowledge point with 'network' mode so as to make technical innovation work.

The invention relates to the technical field of intelligent public transport and especially relates to a candidate bus route calculating method and system and an electronic device. The method comprises the following steps: a) screening a candidate bus stop set according to screening rules, and constructing an initial candidate bus route map according to the screened candidate bus stops; b) carrying out pruning on the initial candidate bus route map to obtain a final candidate bus route map; and c) based on a bus route generation algorithm, calculating a bus route reaching maximum passenger flow within the maximum driving time in the final candidate bus route map to obtain an optimal candidate bus route. The method realizes design of city candidate bus routes based on heuristic search, canquickly find the bus route reaching the maximum passenger flow under the constraint condition of the maximum driving time, fully considers travel requirements of passengers and economic benefit of a bus operation company, and is simple in algorithm parameters, simple in optimization process and high in execution efficiency.

The invention discloses an upstream and downstream intersection traffic flow based heuristic lost road traffic flow estimation method. According to the invention, a connection between an intersectionand roads is established, the flow is marked and roads with unknown flow are determined; an initial flow difference index is calculated; the traffic flow of the roads with the unknown traffic flow iscalculated according to the traffic flow condition of an upstream intersection and a downstream intersection of each road with the unknown traffic flow; flow value update of each road with unknown flow is adopted as a scheme and the flow different index of the scheme is calculated, a scheme having the smallest flow difference is selected for each iteration process until the iteration frequency orthe flow difference index cannot decrease any more, and a mapping set including all road IDs and traffic flow data on a road network is output. According to the invention, the update scheme of each process is calculated according to relation of the inlet traffic flow and outlet traffic flow of the upstream and downstream intersections and heuristic search is performed by adopting the flow different minimization as the target and the iteration frequency as a constraint condition, so that an improved traffic flow distribution condition can be given in a controllable period and a problem of traffic flow data insufficiency is relieved.

The invention provides a method and system for generating combined features of a machine learning sample. The method comprises the steps of (A) acquiring a historical data record, wherein the historical data record comprises multiple pieces of attribute information; and (B) performing feature combination stage by stage on at least one feature generated on the basis of the multiple pieces of attribute information according to a heuristic search strategy so as to generate candidate combined features, wherein a target combined feature is selected from a candidate combined feature set to serve asa combined feature of the machine learning sample in allusion to each stage. According to the method and system, generation of the candidate combined features and selection of the target combined features are accomplished stage by stage through the heuristic search strategy, so that automatic feature combination is effectively realized under the condition of using fewer computation resources, andthe effect of a machine learning model is improved.