40results about How to "Reduce traversal" patented technology

The invention discloses a hybrid cloud storage method based on multi-level cache. The method comprises the following steps: a user joins a user group for login according to the local area network where the user is, and sends a user name and a password for user authentication to a super node of the region where the user group is, the super node achieves authentication of the user according to the received user name and password, the user receives a catalog view returned from the super node and metadata information of all files and sends the catalog view to a group leader node of the user group, the group leader node judges whether the buffer memory space of the group contains files in the catalog view according to the received catalog view, and if yes, feeds the IP address of the node in which the files are saved in the buffer memory space of the group to the user, and the user determines the optimal access path of each file in the catalog view according to the received feedback information. The hybrid cloud storage method allows the user to access data from the user group cache and the super node layer cache which have lower network delay, thereby lowering the network delay of the user data access.

The invention discloses a chain log storage structure, a hash index structure and a data operation method thereof, a server and a medium. The chain log storage structure comprises a plurality of storage segments and an array; a storage unit of each storage segment is a log entry; the log entry includes an initial size field, an index position field, a key length field, a Key field and a Value field, wherein the Value field is used for storing data; the Value field is used for storing data keys; the initial size field is used for storing an initial size of the storage log entry; the index position field is used for storing the position of an index of the log entry in a hash table; and the array is used for storing meta information of each storage segment. The hash index structure comprises a plurality of hash buckets; and a data structure of each hash bucket includes an index number field, a valid mark field, a label field, an offset field and an extension bucket field. According to the chain log storage structure, the hash index structure and the data operation method thereof, the server and the medium, the close continuity of the whole structure memory can be maintained and the fragmentation rate is reduced.

An automatic ortho-rectification method based on a remote-sensing image of an environmental No.1 satellite comprises the following steps of: (1) eliminating projection distortion caused by ground fluctuation in a reference image according to a geometric projection transformation principle to generate a simulated projection image; (2) finding possible feature points in the reference image and an original image by utilizing a Harris corner extraction algorithm; (3) finding possible homonymy points on the basis of the step (2) with a correlation coefficient method; (4) solving a relation coefficient between the reference image and the original image by using a polynomial model based on least squares; and (5) correcting the original image by utilizing the relation coefficient between the reference image and the original image, and eliminating the distortion caused by the ground fluctuation according to the geometric projection transformation principle again, so as to finally finish ortho-rectification. The method has the characteristics of high automation degree, no need of manually selecting control points, high correction precision and high production efficiency.

The present invention discloses an HEVC fast coding method. During the coding process, the temporal similarity between optimal coding units of the same coordinate position in forward and backward reference frames, the spatial similarity between adjacent optimal coding units in the left and right direction, and the spatial similarity between adjacent optimal coding units in the up and down direction, are fully utilized to figure out the depth predicted values of P-frame or B-frame optimal coding units. After that, based on the one-to-one correspondence relationship between the depth predicted values of the optimal coding units and depth ergodic intervals, the depth ergodic intervals of the optimal coding units are determined. The optimal coding units and all coding units in the optimal coding units are coded within the depth ergodic intervals. During the coding process, qualified P-frame or B-frame optimal coding units and qualified coding units are selected quickly in the prediction mode, so that the unnecessary depth traversal and the unnecessary prediction-mode traversal are avoided. Therefore, on the premise that the video quality is ensured, the calculation complexity of the video coding is lowered.

The invention provides a security inspection image-based dangerous article image injection method, and relates to the technical field of security check image processing. The method includes: carryingout image preprocessing on a pre-injected dangerous article image; obtaining a to-be-processed package X-ray image matrix through security inspection equipment; determining whether a package is present, and obtaining the position of the package contour in the matrix; judging whether a dangerous article image can be injected or not according to the sizes of vacancies in the package image, fusing the dangerous article image and the package image if the dangerous article image and the package image can be injected, normalizing the dangerous article image and the package image, generating a picture, and outputting the picture; and normalizing an original package image matrix if the dangerous article image and the package image cannot be injected, and outputting the picture. The method providedby the invention is a method for automatically detecting the injectable position and fusing the image of the dangerous article image in security inspection, and can perform dangerous article image injection on the real-time security check image in a short time, thereby improving the injection accuracy and efficiency, and having better usability and robustness.

The invention relates to the technical field of urban planning, and particularly discloses an efficient OD data visualization method and system, and the method comprises the steps: a dictionary generation module carries out the correlation searching of the OD terminal point information of a corresponding terminal point column or the OD starting point information of a starting point column according to the same OD starting point information in a starting point column of an original data table or the same OD terminal point information in a terminal point column, copying, extracting, permutation and combination are carried out to obtain sub-data tables, and an attribute dictionary is generated according to all the obtained sub-data tables; the data preprocessing module is used for performing data type analysis on the starting point element information and the end point element information respectively, and performing data type conversion on the starting point element information and the end point element information of which the data type is surface data to obtain the starting point element information and the end point element information of which the data type is point data; and the data searching module is used for searching corresponding OD line starting point information and OD line terminal point information from the attribute dictionary. By adopting the technical scheme of the invention, the OD line drawing efficiency can be improved.

The invention discloses an Hbase-based time series data storage method, device and equipment. The method comprises the steps of obtaining a piece of time series data generated by a data source; sorting the plurality of data labels and the at least one monitoring index to obtain a label sorting result and an index sorting result, and associating the label sorting result and the index sorting result with the data names to obtain a sorting relationship; dividing timestamps according to a preset time interval, and recording the timestamps corresponding to the data generation time; forming a row key by the data name, the identification tag and the timestamp corresponding to the data generation time, forming a column of a tag column cluster by the data tags except the identification tag in the plurality of data tags according to a tag sorting result, and forming a column of an attribute column cluster by the at least one monitoring index according to an index sorting result, obtaining a multi-valued model storage structure; and storing the multi-valued model storage structure and the sorting relationship into an HBase database. By applying the method, the data label and the row key can be decoupled, and redundant storage is avoided.

The invention discloses a method for efficiently selecting user communication base stations in a relay cellular network. In a cellular network comprising a central base station and half-duplex and a decode-and-forward relay, the method comprises the following steps of: aiming at all the users, selecting one user which can realize minimization of the power consumption of the system after the serving central base station of the user is adjusted to the relay; under the condition where the central base station and the relay both are not overloaded, adjusting the user to corresponding relay, and keeping the user unchanged, performing the same operation to the rest of users, comparing the power consumption of connection before adjustment with the power consumption of connection after adjustment, and adopting the connection mode of which the power consumption is minimum. The method provided by the invention is suitable for minimizing the power consumption of the whole system under the condition where load of each base station is not high and rate requirement of each user is guaranteed, can efficiently select the communication base stations of the users, and also can reduce total power consumption of the system.

A reliable storage method of a non-document system comprises a writing process of data, a reading process of the data and a deleting process of the data, wherein in the writing process of the data, a local database is opened by an operator, and an offset address of a current write disk is acquired; then a hard-disk device is opened, and a file pointer is deviated to a write position; when the data is written, a write mark is established in the local database, and the data begins to write; in the reading process of the data, the record of the file in the local database is searched according to a search condition transmitted by an application file, an offset address of the file in the hard-disk record is acquired from the record, data is read from the offset address, and the data is returned to the application layer in a callback form; and in the deleting process of the data, the file in the local database is marked as deleted by the operator.

The invention relates to the technical field of scalable video coding, in particular to a spatial scalable fast coding method. The method comprises the following steps of dividing a video sequence into three categories according to a distribution condition in a basic layer; when the coding of an enhancement layer is carried out, computing correlation degree of a current coding unit of the enhancement layer and an adjacent coding unit of the enhancement layer; setting a weight representing the correlation degree for the adjacent coding unit; computing a probability corresponding to each depth of the current coding unit according to the depth and the weight of the adjacent coding unit, and arranging from large to small according to the probability; and predicting the depth of the current coding unit of the enhancement layer according to the category of the video sequence, and removing the depth with a small probability. According to the method, the unnecessary traversing process of the enhancement layer is reduced, and the coding complexity can be reduced.

The invention provides a multi-path cross-class query and optimization method in an object proxy database, which comprises the following steps of: respectively obtaining queries of target attributes by using different path expressions on the same path, and defining the queries as multi-path cross-class queries with the same end point class; for the condition that a plurality of paths in the cross-class queries contain public class nodes, defining the paths as multi-path cross-class queries with different end point classes; then proposing multi-path cross-class query grammar support implementation for supporting multi-path cross-class queries with the same terminal class and different terminal classes; and finally, proposing an execution scheme of the multi-path cross-class querjes, and proposing a multi-path pointer tracking algorithm for the multi-path cross-class queries on the basis according to a pointer tracking algorithm thought used by the original cross-class query, wherein themulti-path pointer tracking algorithm is used for calculating a multi-path expression. By adopting the method and the device, a user can obtain different target attribute expressions by using one path expression, redundant expressions of the common path expression are reduced, and the query efficiency is improved.

The invention relates to the technical field of development auxiliary tools, and provides a tool class calling method and system, a computer device and a storage medium, and the method comprises the steps: identifying a tool class contained in a software program file from a software project file, and obtaining definition information of the tool class; obtaining function identification informationof the tool class according to the definition information; storing the definition information and the function identification information in a development tool template; and when the software projectfile is processed, calling the tool class from the development tool template. According to the tool class calling method, the definition codes of the dispersed tool classes in the software project file can be integrated into the specified development tool template, the tool classes can be called in the development tool template conveniently, the number of the tool classes called after multiple files are traversed and found in the software project file is reduced, and the calling efficiency of the tool classes is improved.