506 results about "Point sequence" patented technology

The invention discloses a gesture identification method and a gesture identification system, wherein the method comprises the following steps: step S1, a gesture detecting step: detecting a gesture in video stream in real time, marking a region in which the gesture is detected as an interested region; step S2, a gesture segmenting step: processing the interested region by utilizing skin color segmentation, and then performing edge detecting and outline extracting to obtain a point sequence of a hand-shaped outline; step S3, a gesture identifying step: firstly extracting a Fourier descriptor of the hand-shaped outline and then mapping the Fourier descriptor to a new vector of a characteristic space by utilizing PCA (principal components analysis), comparing the distance between the new vector and a gesture clustering center obtained by training, and judging a gesture type represented by the vector. The gesture identification method and the gesture identification system disclosed by the invention can improve the gesture identification accuracy and the gesture identification efficiency, and can effectively avoid background color interference.

The invention discloses an intelligent control method of a driverless vehicle tracking a desired trajectory. The method is characterized by comprising the following steps of: establishing a body coordinate system, and determining the position information of the vehicle; calculating a membership degree function of a road curvature characteristic according to a radian value, and defining a fuzzy rule to calculate a decision output speed; according to the distance the vehicle runs for 1.5-2.5 seconds, searching for the trajectory points larger than the distance and closest to the vehicle in the coordinate sequence of the desired trajectory; calculating the moving curvature of the vehicle according to an adaptive proportion-differential-integral control algorithm formula; and finally judging whether the target point of the current control cycle is the last point selected from the road point sequence; if so, outputting a signal to a brake servo control system to start the brake to slow down the vehicle; otherwise, outputting a voltage signal by a controller local area network module to keep and control the accelerator opening so that the vehicle continues proceeding. Through adoption of the intelligent control method, the driverless vehicle can realize a function of tracking the desired trajectory, and has certain adaptability to the change of the road curvature characteristic.

The invention discloses a laser-vision sensing assisted remote teaching method for remote welding, relating to a remote teaching method for remote welding. The purpose of the invention is to solve the problems of difficult-completed remote teaching of complex weld bead in remote welding and long time consumption of position and attitude adjustment of a welding gun in teaching. The method comprises the following steps: 1. initializing a system; 2. setting a weld joint type; 3. automatically obtaining weld bead feature points; 4. calculating to obtain weld bead teaching points Ti = Ps + Rr; 5. recording and forming a teaching point sequence; 6. judging whether the teaching points are enough; 7. downloading a remote teaching program; 8. and finishing teaching. The method is used for remote welding.

Provided is a calibration method for an automotive camera system that can easily calibrate intrinsic parameters and extrinsic parameters of an image taking unit. A calibration method for an automotive camera system 1 includes: recognizing a recognition target object including a preset straight line portion, from an image taken by each of image taking units 307 to 310, and extracting feature points of the recognition target object from the image; projecting the feature points onto a virtual spherical surface 601, estimating intrinsic parameters of the image taking units 307 to 310 on the basis of a shape of a feature point sequence formed on the virtual spherical surface 601, and calibrating the estimated intrinsic parameters; and calculating an overhead point of view of the image on the basis of the feature points, estimating extrinsic parameters of the image taking units 307 to 310 on the basis of the calculated overhead point of view, and calibrating the estimated extrinsic parameters.

The invention relates to a transformer substation inspection robot global path planning method based on magnetic navigation. The method comprises the steps of: modeling an environment of a transformer substation to form a topology network graph and a connecting relationship matrix; determining a searching path matrix, namely the shortest path matrix, by using a Floyd algorithm, and increasing the computing speed by iterative optimization; computing the shortest connecting path between an initial point and a target point sequence by transversing nodes and edges; converting the shortest connecting path into a travel path indicated by an RFID (radio frequency identification device) sequence; and computing the rotating angle of a robot at each RFID at an intersection according to the connectivity of the path. The transformer substation inspection robot adopting the transformer substation inspection robot global path planning method provided by the invention is short in travel path, and rapid to be in place for detection, therefore, the energy of batteries is saved, and simultaneously the time for moving the robot equipment for verification after the operation of the transformer substation equipment can be effectively saved.

The invention discloses a multi-period optimal corner small straight-line segment interpolation method based on S curve acceleration/deceleration control in a numerical control system. The method comprises the following steps of: based on the limitations of the machining precision of a machine tool and the maximum machining speed of each driving shaft, setting the acceleration speed at the joint of the corner and the straight-line segment to be zero according to geometrical parameters of each corner on a machining path, the maximum machining acceleration speed of each driving shaft and an optimization target, and determining and adjusting the optimized corner interpolation parameters in multi-period transition at each corner, so that the acceleration speed at two ends of each small straight-line segment meets the reachability requirement of the S curve acceleration/deceleration control mode; and calculating the time at each acceleration/deceleration stage of each straight-line segment according to the S curve acceleration/deceleration mode, respectively performing straight-line segment and corner interpolation on each small straight-line segment, sequentially outputting interpolation point sequences in real time, and driving a numerical control machine tool to execute the machining operation. By the method, the vibration of the machine tool is reduced, the quality of the surface of a workpiece is improved, the calculation speed is high, the real-time machining requirements can be met, and the method is suitable for different models of three-axis, four-axis or five-axis numerical control machine tools.

Disclosed are a virtual touch panel system and an interactive mode auto-switching method. The virtual touch panel system comprises a projector configured to project an image on a projection surface; a depth map camera configured to obtain depth information of an environment containing a touch operation area; a depth map processing unit configured to generate an initial depth map, and to determine the touch operation area based on the initial depth map; an object detecting unit configured to detect, from each of plural images continuously obtained by the depth map camera after the initial circumstance, a candidate blob of at least one object located within a predetermined distance from the determined touch operation area; and a tracking unit configured to insert each of the blobs into a corresponding point sequence according to a relationship of the geometric centers of the blobs detected in adjacent two of the obtained images.

The embodiment of the invention discloses a floating car track map matching method and a floating car track map matching device, and relates to the field of an intelligent traffic system. The matching error of the floating car track on a map can be reduced, and the matching accuracy can be improved. The method comprises the following steps that firstly, at least one road respectively corresponding to each track point in a track point sequence is obtained, then, the road matching probability between the track points and each road is calculated according to the distance from the track point to each road in the at least one road, next, the road transfer probability from each road corresponding to the track point to each road corresponding to a next track point is calculated, in addition, a plurality of floating car tracks corresponding to the track point sequence are traversed according to the at least one road respectively corresponding to each track point, and finally, the route tracks of a floating car matched on the map are obtained from the floating car tracks according to the road matching probability and the road transfer probability.

The invention provides an all-digital receiving method for a satellite-borne AIS.The method comprises the steps that a signal processed through radio frequency digitization and a local carrier are subjected to orthogonal frequency conversion and then filtered to acquire mixed baseband signals, and the mixed baseband signals are separated; burst detection is performed on separated one-way signals; after a frame header is detected, frequency offset estimation is performed, and frequency offset compensation is performed on the baseband signals on the basis of a frequency offset estimation result; a frame header sequence is selected as an input signal for symbol timing synchronization to acquire a symbol timing error estimation value, and optimal sampling point data is interpolated according to the symbol timing error estimation value; information demodulation is performed on the optimal sampling point data by adopting an incoherent Viterbi demodulation algorithm, a bit information sequence is verified on the basis of a cyclic redundancy check code, and an optimal sampling point sequence of a frame header symbol period and the frequency offset, the amplitude and the initial phase of the verified bit sequence are estimated; a demodulation signal is recovered on the basis of frequency offset, amplitude and initial phase estimation values; the recovered strong signal is subtracted from the separated mixed baseband signals.

A tunnel shield universal segment dynamic 3D virtual assembling system applied in the tunnel technology field comprising a geometry modeling module for setting up the geometry model of the segments and the tunnel axis by parameter design and storing the model into a database, a universal segment selection module for obtaining the geometry model data from the database and processing them to obtain the segment point sequence of the tunnel axis and storing them into an assembling database, a core module namely the universal segment 3D virtual assembling and deviation detection module for extracting the virtual assembly required data from the assembling database interface, displaying the assembling process on the terminal and displaying the deviation between the virtual assembled segment center with the tunnel theoretical axis in the text and graph mode, a universal segment assembling data processing module for executing the data maintenance and communication with the database. The invention realizes the dynamic function of the universal segment in the tunnel shield construction to facilitate the on-time regulation of the tunnel machine posture.

The invention discloses a recognition and counting method for cells, comprising nine steps: preprocessing an image in a micron-order microscopic acquisition environment; extracting cell holes from the preprocessed image; performing closed hole filling of cells by using the knowledge of a connected domain; extracting a contour point sequence of cells from the image filled in step 3; filling non-closed holes of cells by adopting a non-closed hole filling method based on circularity determination; performing chamfer distance transformation on the filled image; performing extreme value uniqueness marking on cell hole positions; segmenting the image after extreme value uniqueness by using a marked watershed method; and quantifying and marking the segmented result. The method has the advantages that the influence of image noise can be greatly reduced, the phenomena of over-segmentation and discontinuous segment lines are eliminated, the segmentation effect is improved, and the cell recognition rate is improved.

The invention provides a signal processing method and device for frequency-agile radar based on a variable repetition frequency technology. The method includes the following steps: designing a frequency point sequence of random hopping according to a random number and the carrier frequency related parameters of radar, and designing a pulse time interval sequence according to the frequency point sequence and the distance range of a target to be detected; emitting a radar pulse signal according to the designed frequency point sequence and the designed pulse time interval sequence, receiving an echo signal of the radar pulse signal, and sampling and preprocessing the echo signal to obtain a baseband echo sampling signal; processing the baseband echo sampling signal by a single pulse to obtaina time-domain echo signal after pulse compression; and coherently accumulating the time-domain echo signal after pulse compression in the two dimensions of frequency point and pulse number, and obtaining the fine resolution distance and Doppler velocity values of the target to be detected through threshold decision. Accurate detection of the distance and velocity of a moving target by coherent frequency-agile radar is realized, and the amount of computation for the processing of coherent frequency-agile signals can be reduced.

The invention relates to a five-axis numerical control processing technique for complex curved surfaces, in particular to a tracking smoothing method suitable for a five-axis numerical control device, which belongs to the technical field of numerical control. The method comprises the following steps of: 1, continuously reading a segment of numerical control command point sequence, and identifying a program segment which satisfies a smooth curve interpolation condition; 2, estimating the positions of interpolation points between adjacent command points, and taking the positions as first-time correction command points; 3, parameterizing the first-time correction command point; 4, generating second-time correction command points, and determining first-order derivative vectors and second-order derivative vectors of the second-time correction command points; and 5, constructing a quintic spline curve for each coordinate component of adjacent second-time correction command points respectively, connecting the quintic spline curves into an integral smooth curve, and performing curve interpolation on the smooth curve. By means of the method, the processing efficiency and the processing accuracy of the five-axis numerical control device are increased, and a turning phenomenon caused by broken line tracks is eliminated.

The invention discloses a transform domain quadratic estimation method uniting weighted threshold de-noising and balanced decision. The method comprises the following steps: step 1), uniting a superimposed training sequence and a harmonic sequence and generating a sending sequence; step 2), in an OFDM (Orthogonal Frequency Division Multiplexing) baseband system, through an estimation method of an LS channel of the superimposed training sequence, obtaining coarse estimation of a channel frequency response; step 3), performing amplitude phase compensation on the coarse estimation of the channel frequency response, adding a window function with a width of M and performing DCT (Discrete Cosine Transformation) of an M point, performing low-pass filtering, time domain de-noising and sequence zero padding extension processing on the obtained sequence, performing N point IDCT (Inverse Discrete Cosine Transform), window removing and secondary amplitude phase compensation on an extended N point sequence, as the following formula, and obtaining the following formula; step 4), performing balanced decision on the obtained result and obtaining the following formula, repeating M point DCT, low-pass filtering, time domain de-noising, interpolation zero padding and N point IDCT, and obtaining a final estimation of a frequency domain response, as the following formula. The transform domain quadratic estimation method uniting the weighted threshold de-noising and the balanced decision provided by the invention eliminates the influence of a data sequence and the training sequence on the channel estimation performance, improves the performance of the existing LS (Least Squares) channel estimation, and has lower implementation complexity.

The invention provides a ship intelligent collision avoidance system based on manoeuvrability modeling, comprising a state sensing subsystem, wherein the state parameters of the ship and the positioninformation of an obstacle are obtained; the maneuverability modeling module processes the ship's own state parameters, constructs the sample pairs, carries on the ship maneuverability on-line modeling, and predicts the ship's possible arrival position at the next time under all feasible maneuverability. The intelligent collision avoidance module combines the position information of the obstacle,Binary navigable area information and collision avoidance rules are used to carry out dynamic path planning. In path planning, the maneuverability modeling module predicts the possible arrival position of the ship at the next time as a constraint, outputs a reasonable planning path point sequence, and decouples it into a heading tracking sequence and a speed tracking sequence. Track the planned real-time heading and speed respectively. The invention realizes the intelligent collision avoidance decision of the ship on the basis of the on-line prediction of the ship maneuverability, and realizesthe safe and autonomous navigation of the ship.

The invention provides a dynamic gesture trace recognition method based on a depth convolution neural network. The method comprises the following steps of collecting an originally input gesture trace point sequence; performing preprocessing ; detecting and eliminating abnormal points of the gesture trace point sequence; performing marginalizing processing on the preprocessed gesture trace point sequence; generating a normalized gesture trace graph; extracting depth features of the normalized gesture trace graph by using a trained depth convolution neural network model; recognizing the shape type of the corresponding gesture trace point sequence by using a trained support vector machine; dividing unknown direction types by a tree classifier according to the shape type of the gesture trace point sequence; fusing the recognized shape type and direction type; and generating a fused trace recognition result of the gesture trace point sequence. The method has the advantages that the shape recognition and the direction recognition are realized; the directed dynamic gesture recognition service is provided for the gesture trace point sequence; the dynamic gesture trace recognition work is not influenced by time and space differences; and the classification is fine.

The invention provides a trajectory tracking method and system for a site unmanned vehicle. The method comprises the following steps that spatial position information of a way point sequence Pi is acquired through a sensor arranged on the site unmanned vehicle, and high-order Bezier curve fitting is carried out on the way point sequence Pi to obtain a trajectory line; delay time is set, a currentposition of the unmanned vehicle is extended in a speed direction, extended distance is a product of speed at a current moment and delay time, an extension line endpoint O(xO, yO) is obtained, therebyfinding a point M (xM, yM) closest to the extension line endpoint O on the way point sequence; a lateral deviation ex(t) and tangent line included angle e(t) between the current site unmanned vehicleand point M are calculated; a current vehicle angle control amount delta(t) is calculated and applied to a lateral control strategy of the site unmanned vehicle. The trajectory tracking method and system for the site unmanned vehicle solves the problem of the dependence of the site unmanned vehicle on inertial navigation equipment, improves the trajectory tracking precision and reliability of thesite unmanned vehicle, and is suitable for the implement of the trajectory tracking of the site unmanned vehicle in different situations.

The invention provides a spraying method and device based on an unmanned aerial vehicle (UAV). The method comprises a track planning step in which a course reversal point R is determined, an area which needs spraying is selected, and a track of a UAV flight point sequence is planned; a spraying step in which flight point sequence flight is traversed successively while spraying is carried out; and a course reversal determining step in which when a predetermined flight point is reached, whether the present pesticide amount and/or electricity amount can satisfy use, if yes, spraying is continued, and otherwise, the course reversal point R is returned to. The spraying device comprises a track planning module, a spraying module and a course reversal determining module. According to the invention, shortage of batteries or pesticide can be discovered timely, the UAV can return automatically to replace the battery or supplement pesticide, time and labor are saved, the efficacy is high, determination is accurate and the spraying efficiency is high.

The invention discloses a path planning method and device, and an inspection robot. The path planning method specifically comprises the following steps: initializing information, determining the quantity of all docking points and all task points, a task starting point, a task ending point, a task list and the task point quantity; reading a shortest distance matrix containing all the docking pointsand all the task pints, a nearest neighbor matrix, the task starting point, the task ending points, the task point quantity and the task list; performing pre-treatment on the task list to obtain a latest task list; calculating the shortest distance matrix according to the latest task list to obtain a current shortest distance matrix corresponding to the latest task list; according to a preset improved ant colony algorithm, performing path calculation on the task points in the latest task list to obtain a task point sequence; performing refining treatment on the task point sequence according to the nearest neighbor matrix and preset one-way road section information to obtain a task point inspection sequence.

The invention discloses a geographic information system (GIS) terminal cable fault on-line location method. The GIS terminal cable fault on-line location method comprises the steps of synchronously acquiring fault current travelling wave signals of three-phase sheaths at two ends of a GIS terminal cable, filtering white noise, pulse noise and high-frequency oscillation noise which are acquired in the fault current travelling wave signals, eliminating outer modulus interference signals and inner modulus interference signals in filtered fault current travelling wave signals, obtaining sampling point sequences corresponding to catastrophe points of the fault current travelling wave signals, identifying the fault current travelling wave signals and performing location according to identified results. The GIS terminal cable fault on-line location method has high applicability, reliability, practical applicability and automation level, and the accuracy of fault location can meet the requirement of underground power cable system fault location.

The invention relates to a face identification based living body detection method and apparatus. The method comprises the steps of: carrying out identity authentication on a to-be-detected user in a to-be-detected crowd through face identification; obtaining a training eye image of the to-be-detected user watching a training sight point, and further extracting training sightline features; obtaining a to-be-detected eye image of the to-be-detected user watching a system random point, and further extracting to-be-detected sightline features; for the to-be-detected user passing the identity authentication, carrying out primary sight point position estimation by adopting an adaptive linear regression algorithm by utilizing a training sight point sequence, the training sightline features and the to-be-detected sightline features; adding the system random point, meeting the predetermined conditions, in the sight point position obtained by primary sight point position estimation into the training sight point sequence to form a new sight point sequence, and carrying out incremental sight point position estimation; and estimating the estimated sight point position according to the incremental sight point position, and carrying out living body judgment by utilizing KL divergence.