617 results about "SIMPLE algorithm" patented technology

The invention relates to a movement human abnormal behavior identification method based on template matching, which mainly comprises the steps of: video image acquisition and behavior characteristic extraction. The movement human abnormal behavior identification method is a mode identification technology based on statistical learning of samples. The movement of a human is analyzed and comprehended by using a computer vision technology, the behavior identification is directly carried out based on geometric calculation of a movement region and recording and alarming are carried out; the Gaussian filtering denoising and the neighborhood denoising are combined for realizing the denoising, thereby improving the independent analysis property and the intelligent monitoring capacity of an intelligent monitoring system, achieving higher identification accuracy for abnormal behaviors, effectively removing the complex background and the noise of a vision acquired image, and improving the efficiency and the robustness of the detection algorithm. The invention has simple modeling, simple algorithm and accurate detection, can be widely applied to occasions of banks, museums and the like, and is also helpful to improve the safety monitoring level of public occasions.

The invention discloses a method for fast scanning and positioning of a matrix keyboard, which relates to a method for the fast scanning and positioning of the matrix keyboard in the field of controlling an embedded system, and is particularly suitable for designing the matrix keyboard of an embedded terminal device. The method comprises the following steps of: initializing a keyboard device and registering the keyboard device; waiting for the keyboard to trigger an interruption signal by a system; triggering a keyboard interruption signal; responding to the interruption signal by the system; performing keyboard scanning; judging whether keys are effective; and outputting a keyboard key value to a queue, and waiting for an application program to read the key value. The method can effectively avoid instable and dithering levels of the matrix keyboard because of per se mechanical properties, improves the accuracy of the scanning and the positioning of the matrix keyboard, and has the characteristics of simple algorithm, steady performance, high control accuracy and the like.

The invention discloses a spatial circle geometrical parameter dual-eye solid visual measurement method suitable for non-contact type online measurement of the spatial circle geometrical parameter of an industrial product, aiming at overcoming the existing problems of low measurement precision, slow measurement speed and low automation degree. The method is divided into a calibrating stage, an image processing stage and a spatial circle fitting stage. A neural network technique is firstly adopted to carry out the calibration of a camera; a dual-camera is utilized to carry out the extraction of sub-pixel image edges of the spatial circle, a simple algorithm is developed based on the image gray gradient distribution characteristic so as to realize the match of edge points, the practical 3D spatial coordinates of the circle edges are then obtained, and the geometrical parameters of the spatial circle comprise radius, position of center of circle and orientation of the plane where the spatial circle is arranged are gained by the fitting of the spatial circle with corresponding mathematical geometrical knowledge. The spatial circle geometrical parameter dual-eye solid visual measurement method has fast image processing speed and high automation degree, and has relative measurement error of the spatial circle of superior to plus or minus 0.6% when the plane where the spatial circle is arranged has an angle of about 50 DEG with the image plane.

The invention discloses a multisource information fusion method in an evidence high-conflict environment, which comprises the following steps of: determining the mutual support degree of evidences according to evidence distances, further calculating the relative discount factor of each evidence by taking a characteristic vector corresponding to the maximum characteristic value of a evidence support degree matrix pattern as a weighting vector of the evidence, wherein the evidence with the maximum weighing coefficient is a key evidence; and discounting each evidence information, and fusing the information of multisource sensor data by using the D-S (Dempster/Shafer) rule so as to increase the accuracy and precision of a monitoring proposition judgment under an evidence high-conflict status.The method can effectively eliminate the uncertainties of multisource information, increase the accuracy of information evaluation and overcome the limitation that correct evaluation results can not be acquired by using the traditional D-S evidence theory under the evidence high-conflict status, and has the advantages of simple algorithm, high convergence rate and the like.

The invention relates to a three-dimensional model reconstruction method of tree point cloud data based on partition and automatic growth, comprising the following steps: preprocessing; estimating the main curvature of point cloud; partitioning data according to the main curvature; calculating a framework of a principal branch of branches by the partitioned point cloud belonging to a branch part;leading from the framework of the principal branch of the branches to generate twigs of the branches by the partitioned point cloud belonging to a leaf part; and generating a branch mesh model and adding a leaf model at the ends of the twigs of the branches. In the method, the three-dimensional reconstruction model which is faithful to original entities can be obtained by means of the scanning data of a laser scanner; the reconstruction model of the tree point cloud data can be obtained based on data partition and growth of the twigs of the branches, thus the method has simple algorithm and accurate calculation result; and the calculation result has important application value in the fields such as virtual reality, computer games, natural scene simulation, urban landscape design, film production, three-dimensional reconstruction of trees, agriculture and forestry measurement and the like.

The invention relates to a non-overlapping visual field multiple-camera human body target tracking method, in particular to a method for carrying out target matching and tracking in multi-camera videos of a non-overlapping visual field under the condition of known topological relation among multiple cameras, comprising the following steps: firstly, establishing a topological relation matrix among multiple cameras, then extracting a foreground image of a pedestrian target followed by selecting a tracking target in the visual range of a certain camera by a human-machine interaction interface, using a mixed Gaussian model to extract the appearance characteristics of the target, comprising colour histogram characteristic and UV colour characteristic, in the whole tracking process, adopting a tree structure to manage tracking task, using Bayesian model to estimate overall similarity, and searching for the optimal matching target according to the overall similarity. The invention has the advantages of high precision, simple algorithm, easy realization, strong instantaneity and the like, and can provide a new real-time and reliable method for multi-camera target recognition and tracking.

The invention provides a keyframe extracting method and system for monitoring system videos. The method comprises the following steps of (a) utilizing a background differencing method to extract keyframe sequences including movement objects in the videos; (b) calculating the similarity degree of every two adjacent frames in the keyframe sequences based on a joint histogram, extracting keyframes with the similarity degree smaller than a threshold value and adding the keyframes into a keyframe set; (c) judging whether mark numbers of every two adjacent keyframes in the keyframe set are smaller than a certain interval or not, deleting the keyframes with small information entropies and updating the keyframe set if the mark numbers are smaller than the certain interval. By utilizing the keyframe extracting method and system for monitoring the system videos, the keyframes including the movement objects can be extracted from a large amount of monitored videos by using a simple algorithm, redundant data in the keyframe sequences are remarkably decreased, and the storage amount of video data are remarkably decreased.

The invention discloses a three-dimensional gesture estimation method and a three-dimensional gesture estimation system based on depth data. The three-dimensional gesture estimation method comprises the following steps of S1, performing hand region of interest (ROI) detection on photographed data, and acquiring hand depth data, wherein the S1 comprises the processes of (1), when bone point information can be obtained, performing hand ROI detection through single bone point of a palm; (2) when the bone point information cannot be obtained, performing hand ROI detection in a manner based on skin color; S2, performing preliminary estimation in a hand three-dimensional global direction, wherein the S2 comprises the processes of S21, performing characteristic extracting; S22, realizing regression in the hand global direction according to a classifier R1; and S3, performing joint gesture estimation on the three-dimensional gesture, wherein the S3 comprises the processes of S31, realizing gesture estimation according to a classifier R2; and S32, performing gesture correction. According to the three-dimensional gesture estimation method and the three-dimensional gesture estimation system, firstly cooperation of two manners is utilized for dividing hand ROI data; afterwards estimation in the hand global direction is finished according to a regression algorithm based on hand ROI data dividing; and finally three-dimensional gesture estimation is realized by means of the regression algorithm through utilizing the data as an aid. The three-dimensional gesture estimation method and the three-dimensional gesture estimation system have advantages of simple algorithm and high practical value.

The invention discloses a personal identity identification method based on hand shapes, which adopts an imaging device to obtain human hand images, and obtaining personal character description through computer analyzing and processing, and achieving personal identity identification. The concentrate implementation step comprises the following steps: firstly, adopting an imaging device to obtain hand shape images, secondly, extracting hand shape outlines on the hand shape images, thirdly, extracting hand shape characteristic points according to the hand shape outlines, fourthly, forming hand shape characteristic vectors according to the hand shape characteristic points, fifthly, matching hand shape characteristics according to the hand shape characteristic vectors. The personal identity identification method is mainly characterized in of simple algorithm, high received degree of a tester and stable performance, which is easy to achieve multi-modality identification. The personal identity identification method is particularly suitable for the occasions of a gate inhibition and the like.

The invention discloses an optic disc projection location method synthesizing vascular distribution with video disc appearance characteristics, which comprises the following steps of: (1) extracting an interest retina fundus image region by use of a mask operation; (2) carrying out normalization enhancement on a fundus image based on an image observation model; (3) achieving extraction and segmentation of the fundus image by using a non-vascular structure inhibition operator and combining with a hysteresis multi-threshold processing technology; (4) setting a vertical window of double main blood vessel width to slide on a blood vessel segmentation image along the horizontal direction and calculating a vascular distribution degree D(x) at each horizontal position x to obtain a distribution degree curve of a horizontal projection, wherein the minimum value point of the curve is determined to be optic disc horizontal coordinate xod; (5) setting a rectangular window side length of which is equal to the optic disc diameter to slide up and down at the horizontal coordinate xod, estimating local region brightness IN(xod, y) and edge gradient information gN(xod, y) respectively, drawing a vertical projection curve reflecting a change of a characteristic value f(y)=IN(xod, y)*gN(xod, y), wherein a maximum value point of the curve is optic disc vertical coordinate yod. The method has a simple algorithm, a high success rate and excellent robustness.

The invention relates to a vibration monitoring structure based on optical fiber polarized light time domain reflection sense, comprising a laser (1), a first coupler (2), a second coupler (7), a third coupler (8), a fourth coupler (11), a fifth coupler (13), an acoustic optical modulator (3), a polarizer (4), a circulator (5), a sensing optical fiber (6), a first analyzer (9), a second analyzer (10), a first detector (12), a second detector (14), a first low-pass filter (15), a second low-pass filter (16), a first analog-to-digital converter (17), a second analog-to-digital converter (18), anFPGA (19) and a microprocessor MCU (20); the method comprises the following steps: measuring the changes of the polarization state of each optical fiber section with time according to a plurality ofmeasurements, combining Fourier transformation technology to obtain frequency spectrum information, and analyzing the frequency spectrum information to obtain vibration information. Compared with theprevious method which is adopted by a POTDR system and carries out iteration operation on polarization state, the method has the advantages of simple algorithm and small data operation and greatly reduces monitoring time and cost of the system.

The invention discloses a digital background calibration circuit used for a high-speed and high-precision pipelined analog-to-digital converter. The digital background calibration circuit comprises a pseudo random number generator, pipelined circuits with calibration functions and a digital background calibration engine. Based on the structure of the traditional pipelined analog-to-digital converter, a primary-stage pipelined circuit and a secondary-stage pipelined circuit are modified in the circuit disclosed in the invention to realize the injection of random signals and the digital background calibration engine is used to associate the random signals so as to realize the real-time extraction and compensation of error information, thereby avoiding the influence of irrational factors of the traditional pipelined analog-to-digital converter (such as capacitor mismatching, limitations of operational amplifier gains and the like) on the conversion precision of the analog-to-digital converter. The technology can lower the design difficulty of an analog circuit and ensure the performance of a system. At the same time, because of the simple algorithm and the low implementation complexity, the calibration circuit can be used to effectively reduce the area of the chip and lower the power consumption of the system, thereby being especially applicable to a high-speed system.

The present invention provides a binocular vision depth feature and apparent feature-combined face living body detection method. The method includes the following steps that: step 1, a binocular vision system is established; step 2, a face is detected through the binocular vision system, so that a plurality of key points can be obtained; step 3, a binocular vision depth feature and a classification score corresponding to the binocular vision depth feature are obtained; step 4, a complete face area is intercepted from a left image, the complete face area is normalized to a fixed size, and a local binary pattern (LBP) feature is extracted so as to be adopted as an apparent feature descriptor; step 5, a face living body detection score corresponding to a micro-texture feature is obtained; and step 6, the classification score corresponding to the binocular vision depth feature obtained in the step 3 and the face living body detection score corresponding to the micro-texture feature obtained in the step 5 are fused in a decision-making layer, so that whether an image to be detected is a live body can be judged. The binocular vision depth feature and apparent feature-combined face living body detection method of the invention has the advantages of simple algorithm, high operation speed, high precision and the like. With the method adopted, a new and reliable method can be provided for living body face detection.

The invention discloses an image cross reconstruction-based single-sample registered image face recognition method. The method adopting the image cross reconstruction realizes the personal identity authentication by computer analysis and processing, and particularly comprises the following steps: registering low-frequency information of standard face images in a face database; acquiring a front image of a face to be recognized; extracting high-frequency information of the image of the face to be recognized; performing the image cross reconstruction on the high-frequency information of the image of the face to be recognized and the low-frequency information of the registered face images; and finally, performing matching judgment on the fused images and the image of the face to be recognized to realize the identity recognition. The image cross reconstruction-based single-sample registered image face recognition method has the advantages of small space of the database registering the face information, simple algorithm, short recognition time, high recognition rate and the like.

In the field of user authentication, the present invention provides an integrated system for the mutual authentication of a system database and a registered user with a view to increasing the security of remote authentication and the prevention of “phishing/man-in-the-middle” attacks, by one of several alternative means including Code matching, PIN verification, Image reproduction and recognition, Signature and personal data verification, DNA verification and Biometric verification, in each case by means of the differential between variable Codes computed at the database from data recorded for that user and at a remote terminal from replicate data retrieved from a data carrying device. The Codes are derived from the recorded data and a simple algorithm such that the Codes are not predicable.

The invention relates to a method for detecting and classifying glass defects based on machine vision. The method comprises the following steps of: extracting a defect area in a picture provided by a camera (linear scanning) by Canny edge detection to obtain a minimum connected domain of the defects; processing a target area according to a filter and a W characteristic which are provided by the method; defining nine kinds of characteristic modes, scanning the minimum connected domain according to rows and columns, and counting the emerging frequency of the nine kinds of characteristic modes in a sample; and judging the types of the defects (bubbles are used as hollow defects, and impurities are used as solid defects) on the basis of the emerging frequency. Compared with the prior art, themachine-vision-based method has the advantages of simple algorithm, quick operation speed, high accuracy and the like, and a new reliable method is provided for the detection of the glass defects.

The invention provides a network topology layout method and the network management device thereof. And the method comprises: obtaining connection relation between nodes, where the nodes include common nodes and gateway nodes; according to the connection relation, obtaining numbers of times of shortest path routing from common nodes to the nearest gateway nodes; calculating node coordinates and making the common nodes with the same number of times of shortest path routing compose an array, where the difference value between the numbers of times of shortest path routing between the adjacent arrays is equal to 1; displaying the common nodes and drawing the connection relation between the common nodes. And the invention has the advantages of simple algorithm, low cost, and clear layers of network topology and clear network topology.

A system and method in which query load balancing process is performed in a cluster of middle-tier computing elements according to the load on at least one, but more preferably a plurality, of the computing elements of the group, this by passing pointers to queries between machines with information or processes pertinent to the query. Optionally the balancing may be carried out in two stages. The first stage performed rapidly, preferably implemented as a hardware device such as a switch for example in which a plurality of queries is distributed to a plurality of computing elements according to at least one descriptor attached to each query or a simple algorithm is used. The second stage preferably performed as above by at least one computing element of the group, but may optionally be performed by a plurality of such elements. Alternatively, the second stage may be performed by a separate computer.

The invention discloses a memory management method applicable for an embedded system, comprising steps of: providing a memory area with a certain volume; dividing a memory area into a plurality of memory branch areas and unit memory blocks; when it is necessary to provide memory, distributing the unit memory block greater than needed memory for use. The memory management method related to the invention uses a memory layout aiming at the embedded system, distributing memory via simple algorithm in a high effect manner and detecting whether memory error occurs.

The invention relates to the field of data encryption, and discloses a method and a device for encrypting by performing data element division on data, performing coordinate distribution on each data element and adopting different algorithms and keys. The device is characterized in that: the security guarantee of the data does not depend on the complexity of a certain encryption algorithm or the length of a key for encrypting any more. By introducing various custom simple algorithms which cannot be predicted outside and mixing various different encryption algorithms to encrypt the same document, the advantages of the algorithms are effectively integrated, the computational load and cost for encrypting are greatly reduced, and the method and the device have significant beneficial effects and technical progress.

A non-volatile memory has its cells' thresholds programmed within any one of a first set of voltage bands partitioned by a first set of reference thresholds across a threshold window. The cells are read at a higher resolution relative to a second set of reference thresholds so as to provide additional soft bits for error correction. The reference thresholds of the second set are set up to be non-uniformly distributed on the threshold window so as to provide higher resolution at designated regions. At the same time they are conducive to be read in groups for soft bits to be read bit-by-bit systematically with a simple algorithm and read circuit and using a minimum of data latches. This is accomplished by relaxing the requirement that the first set of reference threshold is a subset of the second set and that the resulting soft bits are symmetrically distributed about the hard bits.

The present invention discloses a non-air conduction speech enhancement method based on a multi-band spectral subtraction. Because of the noise in the radar-based non-air conduction speech is always colored and no uniform influence on the speech signals within the whole range of the frequency spectrum, the method divides the speech frequency spectrum in a targeted manner into five sections without overlapping; simultaneously, each section is provided with an individual spectral subtraction coefficient, so as to achieve the effectiveness and pertinence of the algorithm. The embodiment proves the non-air conduction speech enhancement method can effectively compensate the weakness of low pertinence in the traditional speech enhancement method; moreover, the method has the advantages of highly efficient implementation, simple algorithm and obvious effect. Therefore, the method has higher practical value and application prospects.