34 results about "Quicksort algorithm" patented technology

The invention discloses a collaborative Web service performance prediction method based on position clustering. The method comprises the following steps: according to position information of users and services, respectively clustering the users and services which are the same as an Autonomous System; calculating a centroid-based similarity; using a quick sorting algorithm to perform descending sort on similar clusters, finding the previous K neighbor clusters as candidate clusters, so as to finish the neighbor pre-selection; calculating the similarity between an active user and a user in each user candidate cluster, and the similarity between an active service and each service candidate cluster; respectively using the previous K most similar users to predict a performance value of an unknown service for the active user and using the previous K most similar services to predict a performance value of the unknown service for the active service; and synthesizing all performance values to calculate a quality predicted value of a source user relative to a target service. The collaborative Web service performance prediction method based on the position clustering provided by the invention can solve the problems like data sparse and cold boot in Web service recommendation, and improve the precision and coverage rate of Web service performance prediction and Web service recommendation.

A high precision target sphere center extraction method in an unstructured environment provided by the invention belongs to the field of binocular vision technology. According to the invention, denoising is performed based on median filtering and deletion of small area noisy points on an image obtained through photographing and pretreatment such as binarization processing is carried, so that an outline point set of a high precision target sphere center in the image is acquired. By adopting a quick sort algorithm, sorting of the outline point set specific to y coordinates is performed, so that a valid fit pint set used for oval fitting on the outline of the high precision target sphere with non-uniform surface gray level is acquired. By adopting an oval fitting algorithm, the coordinate value of the outline center of the high precision target sphere with non-uniform surface gray level is calculated. The method is high in extraction precision and is used for subsequent operation such as camera calibration, image processing and the like. Especially, the precision of a calibration method utilizing the high precision target sphere is improved and improvement on visual measurement precision is realized in a specific degree.

The invention relates to a piracy tracing watermarking method applicable to streaming media environment, and belongs to the multimedia information safety field. The method comprises the following steps: firstly segmenting video before the watermark embedding procedure; according to content of streaming video, selecting macroblocks meeting the conditions to be embedded with watermarks during the watermark embedding process; and determining the macroblock which is best suited for being embedding with watermark information by adopting a quicksort algorithm. In the piracy tracing watermarking method, based on difference between horizontal direction and vertical direction of motion vector of the macroblock, appropriate motion components are adaptively selected to be embedded with 2-bit watermark information, and the NAL (network abstraction layer) head of a B-frame is modified after the watermark embedding procedure so that the watermark can cope with the situations such as frame loss and the like resulting from network transmission. By adopting the piracy tracing watermarking method, the watermark information can be embedded into the streaming video in real time so as to achieve the purpose of piracy tracing.

The invention relates to an application method for performing satellite-borne SAR (Synthetic Aperture Radar) image ortho-rectification based on a DSM (Digital Surface Model) of median filtering, wherein the median filtering is a type of nonlinear smoothing filtering, depends on a fast sorting algorithm, and is a domain operation. The application method comprises the following steps of firstly determining a domain which takes a certain pixel as a center point, sorting the gray value of each pixel in the domain, and taking the median value as a new gray value of the center point pixel, wherein the domain is called as a window; after the window is moved up and down as well as left and right, a smoothing processing is performed on an image by utilization of a median filtering algorithm. By employment of the median filtering method, the smoothing filtering processing is performed on the DSM of an optical satellite image, and the distortion phenomenon is eliminated.

The invention relates to an image median filtering method based on an incomplete rapid sorting algorithm. The image median filtering method includes the following specific steps that according to a row-major sequence or a line-major sequence, all pixels of an image are scanned and all pixels in a certain neighbourhood of a target pixel of the image are obtained; an element in the sequence is randomly selected as a pivot point, assuming that the position of the pivot point after one-time scanning processing is changed to be m, all the elements which are in an image pixel sequence and are smaller than the pivot point value are moved to be in front of the pivot point and all the elements which are in the image pixel sequence and are larger than the pivot point value are moved to be in rear of the pivot point, and an original sequence is divided into two parts; the position of the pivot point is judged after rapid sorting, if the position of the pivot point is just the middle position of the sequence, a k value is obtained, the K value is used for replacing the value of the image target pixel, and the process is terminated; otherwise, according to the size relation between m and k, a proper subsequence is chosen to repeat the process. According to the image median filtering method based on the incomplete rapid sorting algorithm, a large amount of data sorting and time-consuming data moving operation is reduced and the processing speed is greatly improved when the processing precision is ensured.

The invention discloses a Chinese news story segmentation method based on flexible semantic similarity measurement. The method includes the following steps that a target text set is input, and word segmentation is conducted on news storyboards Ti in the text set; a context relation diagram is built; iteration spreading is conducted on context relevancy between words through the context relation diagram and a rapid sorting algorithm to acquire a flexible semantic relevant matrix; flexible semantic similarity between sentences is defined through the flexible semantic relevant matrix; a Chinese news story is segmented through the flexible semantic similarity. By the adoption of the flexible measurement method, the semantic similarity between the words and between word sets can be more reasonably expressed. Experiments show that compared with a traditional similarity measurement method, in a Chinese news story segmentation technology and based on same segmentation principles, the flexible semantic similarity measurement method can improve segmentation accuracy to 3% to 10%.

The invention relates to a piracy tracing watermarking method applicable to streaming media environment, and belongs to the multimedia information safety field. The method comprises the following steps: firstly segmenting video before the watermark embedding procedure; according to content of streaming video, selecting macroblocks meeting the conditions to be embedded with watermarks during the watermark embedding process; and determining the macroblock which is best suited for being embedding with watermark information by adopting a quicksort algorithm. In the piracy tracing watermarking method, based on difference between horizontal direction and vertical direction of motion vector of the macroblock, appropriate motion components are adaptively selected to be embedded with 2-bit watermark information, and the NAL (network abstraction layer) head of a B-frame is modified after the watermark embedding procedure so that the watermark can cope with the situations such as frame loss and the like resulting from network transmission. By adopting the piracy tracing watermarking method, the watermark information can be embedded into the streaming video in real time so as to achieve the purpose of piracy tracing.

The invention discloses a collaborative Web service performance prediction method based on position clustering. The method comprises the following steps: according to position information of users and services, respectively clustering the users and services which are the same as an Autonomous System; calculating a centroid-based similarity; using a quick sorting algorithm to perform descending sort on similar clusters, finding the previous K neighbor clusters as candidate clusters, so as to finish the neighbor pre-selection; calculating the similarity between an active user and a user in each user candidate cluster, and the similarity between an active service and each service candidate cluster; respectively using the previous K most similar users to predict a performance value of an unknown service for the active user and using the previous K most similar services to predict a performance value of the unknown service for the active service; and synthesizing all performance values to calculate a quality predicted value of a source user relative to a target service. The collaborative Web service performance prediction method based on the position clustering provided by the invention can solve the problems like data sparse and cold boot in Web service recommendation, and improve the precision and coverage rate of Web service performance prediction and Web service recommendation.

The invention relates to a mass data comparison method and system. The method comprises the following steps that: obtaining a mass data file to be compared; according to a pre-stored quick sorting algorithm, carrying out sorting processing on the mass data file to obtain a plurality of sub data files; and carrying out data consistency comparison on the plurality of sub data files. By use of the mass data comparison method and system provided by the invention, the mass data file is divided into the plurality of sub data files, the plurality of sub data files are independently compared so as to effectively solve an existing phenomenon that a single machine has limited I / O (Input / Output) and can not carry out multi-thread comparison, and data comparison efficiency can be improved.

The invention provides an account checking method and system for a unified mobile payment platform, a storage medium and a terminal, and the method comprises the following steps: obtaining bill data and transaction flow data corresponding to each payment order in a preset time interval, the bill data and the transaction flow data containing account checking identification information corresponding to the payment order; sorting the transaction flow data by adopting a quick sorting algorithm; for each piece of bill data, searching transaction flow data matched with the account checking identification information in the sorted transaction flow data by adopting a binary search algorithm; if the searching is successful, marking the matched transaction flow data as successful account checking; otherwise, marking the bill data as an abnormal bill; and traversing the transaction flow data, and marking the transaction flow data which is not marked as successful account checking as an abnormal bill. According to the account checking method and system of the unified mobile payment platform, the storage medium and the terminal, rapid account checking of the unified mobile payment platform can be realized, the time complexity of the account checking process is greatly reduced, and the space complexity is not increased.

The invention discloses a metamaterial simulation data processing method and a metamaterial simulation data processing device. The metamaterial simulation data processing method includes the steps of obtaining N transferring samples through a simulation algorithm, wherein each transferring sample comprises structural parameters characterizing metamaterial units and N is a natural number, obtaining distance values respectively comparing electromagnetic response characteristics of metamaterial units expressed by the N transferring samples with an electromagnetic response characteristic of a metamaterial unit expressed by a target sample, sequencing the N obtained distance values according to a fast sequencing algorithm, obtaining the transferring samples corresponding to the N distance values which are smaller in the sequenced distance values, and treating the transferring samples corresponding to the N distance values as designed samples, wherein n is a natural number less than N. The metamaterial simulation data processing method and the metamaterial simulation data processing device has the advantages of being capable of achieving automatic processing of metamaterial simulation data and improving data processing efficiency.

The invention relates to a converter voltage-sharing method based on nearest level approximation. The converter voltage-sharing method comprises the steps of: calculating a number N(t) of sub-modulesthat need to be triggered at a triggering moment by adopting a nearest level approximation modulation strategy, and acquiring a capacitor voltage sequence; setting a sub-module voltage-sharing recursive threshold RT, so as to determine a sorting method used by the voltage-sharing strategy at different numbers of recursive times; starting sub-module capacitor voltage recursive sorting, counting a number of recursive times, adopting a rapid sorting algorithm for sorting the sequence if the number of recursive times is less than a recursive threshold, otherwise, adopting a heap sorting algorithmfor sorting; and acquiring a sub-module capacitor voltage sequence after sorting, determining a current direction of a bridge arm, selecting the N(t) sub-modules with the lowest capacitor voltage to be put into charge if the current direction is a flow-in direction, and selecting the N(t) sub-modules with the highest capacitor voltage to be put into discharge if the current direction is a flow-outdirection.

The present application discloses a data information sorting method, which includes: a CPU side executes a main program algorithm separated from a quick sort algorithm; wherein, the process of the CPU side executing the main program algorithm includes transferring a data set to be sorted to an FPGA terminal, and sends a sorting start control command to the FPGA terminal; after the FPGA terminal obtains the sorting start control command, it executes the kernel program algorithm separated from the quick sorting algorithm, so that the data set to be sorted is analyzed. Perform sorting processing to obtain the corresponding sorting results. By adding an FPGA terminal to run the core program of the sorting algorithm, the invention breaks through the original hardware limitation, increases the operation speed and improves the work efficiency. The invention also provides a data information sorting system, a CPU end, and an FPGA end, which can run the data information sorting method, and can also increase the operation speed and improve the work efficiency.

The invention relates to a quick paging sorting method based on a quick sorting algorithm. It is difficult for the existing sorting method to know which data are included in the specified range and the order between them before completely sorting all the data, so the existing technology is to completely sort all the data first, and then output the data subset of the specified range, which is There must be a large amount of unnecessary computing overhead, which affects the response speed of the computer system. The method described in the present invention mainly adopts a divide-and-conquer strategy similar to the Quick Sort algorithm, utilizes the cache mechanism, continuously excludes elements that do not belong to the specified range, and gradually sorts and locates the data within the specified range, thereby solving the problem of Problem with quick paging sort. The method of the invention can effectively solve the problem of paging and sorting under the massive data environment, greatly reduce unnecessary calculation overhead, avoid repeated calculation, and improve the response speed of the computer system.

The invention discloses a metamaterial simulation data processing method and a metamaterial simulation data processing device. The metamaterial simulation data processing method includes the steps of obtaining N transferring samples through a simulation algorithm, wherein each transferring sample comprises structural parameters characterizing metamaterial units and N is a natural number, obtaining distance values respectively comparing electromagnetic response characteristics of metamaterial units expressed by the N transferring samples with an electromagnetic response characteristic of a metamaterial unit expressed by a target sample, sequencing the N obtained distance values according to a fast sequencing algorithm, obtaining the transferring samples corresponding to the N distance values which are smaller in the sequenced distance values, and treating the transferring samples corresponding to the N distance values as designed samples, wherein n is a natural number less than N. The metamaterial simulation data processing method and the metamaterial simulation data processing device has the advantages of being capable of achieving automatic processing of metamaterial simulation data and improving data processing efficiency.

The invention belongs to the technical field of integrated circuit design, and specifically relates to a method for dividing a parallel GPDT algorithm on a multi-core SoC. The parallel GPDT algorithm includes two iterations, the inner iteration is responsible for solving the working set, and the outer iteration is responsible for updating the working set. In terms of the critical path of calculation speed, the critical path of the outer loop is the update of the gradient, and the critical path of the inner loop is the calculation of the vector after each projection. These two parts of matrix operations need to be processed in parallel on multi-core; while the rest The calculations can only be implemented serially on the main core, including the gradient projection operation implemented by the Dai-Fletcher algorithm, and the update of the working set by introducing the quick sort algorithm. The vector obtained after the calculation is the support vector of the training data of the GPDT algorithm.

The invention discloses a copper plate packaging and packaging method, and relates to the technical field of copper plate inclusion methods. The method includes the following steps: read out the relevant data of the copper plate from the database, and use an integration algorithm to process the above data; use a quick sort algorithm to process the data processed by the integration algorithm, and the robot completes each step according to the processed data. Packing per pack, per order, and batch of copper plates. The method sorts and packs the data for the integration algorithm, the quick sort algorithm, and the mixed packet algorithm. The operation of the data field is mainly for the processing of the data boundary to prevent data from being reused and unused, and then transmit it to the robot in a specified format The function of automatic packaging is realized, which improves the work efficiency of packaging.

A method, device and equipment for data storage, including: starting a high-speed Huffman Huffman compression algorithm including a quick sort algorithm to perform first data compression on written data to generate a first compressed file; the first compressed file write After entering the Hbase database table, call the double compression algorithm to realize the second data compression of the first compressed file, and output the second compressed file; based on the second compressed file, realize Hadoop distributed file system HDFS data storage. After adopting the embodiment of the present invention, the IO pressure of the network and the disk can be reduced, and the performance of data storage can be ensured.