38 results about "Block graph" patented technology

The invention discloses an RADAR parallax image optimization method, a stereo matching parallax image optimization method and a stereo matching parallax image optimization system. The RADAR parallax image optimization method comprises the following steps: obtaining a color block graph: performing contrast enhancement on an initial image, converting the initial image from an RGB space into a CIELab space, and performing color blocking on the CIELab space by mean-shift color partitioning to obtain the color block graph; acquiring edge information of a parallax image: receiving an initial parallax image of the initial image, and extracting the edge information of the parallax image in the initial parallax image by combining Canny operator; optimizing the parallax image: performing inconsistent region detection on the color block graph and the edge information of the parallax image to obtain a problem region graph, performing OccWeight correction on the initial parallax image according to the problem region graph, and filtering to obtain a final parallax image. According to the method disclosed by the invention, after the initial parallax image is optimized, the error rate is reduced, and the accuracy of the final parallax image is improved.

The invention discloses an encrypting file system based on an EXT2 file system. The encrypting file system based on the EXT2 file system comprises a plurality of block groups, and each block group comprises memory spaces corresponding to the data of a superblock, a block group description list, a block bitmap, a file and catalogue index node bitmap, a file and catalogue index node list and a plurality of data blocks. A plurality of block group description lists are stored in a memory space corresponding to the block group description list. Each block group further comprises a secret key index node bitmap which is arranged behind the file and catalogue index node list and a memory space corresponding to a secret key index node list. The memory space corresponding to the block group description list is further provided with a file system digital certificate (DC). By means of the encrypting file system based on the EXT2 file system, the security of the EXT2 file system is improved. The invention further provides an encrypting method and a deciphering method based on the EXT2 file system.

There are provided an environment recognition device and an environment recognition method. An exterior environment recognition device obtains an image in a detection area, generates a block group by grouping, based on a first relative relationship between blocks, multiple blocks in an area extending from a plane corresponding to a road surface to a predetermined height in the obtained image, divides the block group into two in a horizontal direction of the image, and determines, based on a second relative relationship between two divided block groups, whether the block group is a first person candidate which is a candidate of a person.

The embodiment of the invention provides a data transmission method and device which are used for Ethernet data retransmission to at least solve the problem that an existing Ethernet link cannot ensure reliable data transmission, and relates to the field of communication. The method comprises the steps that a first block group sent by the sending end is received, wherein the first block group comprises multiple blocks and a first retransmission block, and the first retransmission block comprises a sequence field and a verification field; a second block group is generated, wherein the second block group comprises multiple blocks and a second retransmission block, the second retransmission block comprises a confirmed or unconfirmed field, and the confirmed or unconfirmed field comprises a serial number subfield and a confirmed or unconfirmed identifier subfield, a value of the serial number subfield is a serial number of the first block group, when a verification result of the first block group is correct, a value of the confirmed or unconfirmed identifier subfield is a confirmed identifier, and when the verification result of the first block group is incorrect, the value of the confirmed or unconfirmed identifier subfield is an unconfirmed identifier; the second block group is sent to the sending end.

The invention provides a farmland identification method and system, and an unmanned aerial vehicle applied to agriculture, which are applicable to the technical field of agriculture. The identification method comprises the following steps of acquiring a satellite image; segmenting the satellite image to obtain a segmentation block group, wherein the segmentation block group at least comprises morethan two segmentation blocks; removing a non-farmland block from the segmentation block group, and obtaining a residual segmentation block group, wherein the residual segmentation block comprises atleast one segmentation block; identifying a farmland block from the residual segmentation block group based on vegetation index feature. The invention is advantageous in that firstly, the satellite image is segmented, and non-farmland blocks are preliminarily removed from the segmented block group, and then the farmland blocks are further identified from the residual segmentation block group basedon the vegetation index feature; the implementation process is simple and recognition accuracy is quite high.

The invention provides a foresight anti-lock dynamic allocation method for multi-mobile robot system grid points based on a grid map. According to the method, when the grid points are allocated, eachgrid point is only allocated to one mobile robot at most, and collision between the robots can be effectively avoided. Each robot only can obtain preset NC grid points per round at most, the phenomenon that a lot of robots in the system wait for a long time is avoided, the aim that the robots share the grid map is achieved, and the overall efficiency of the multi-robot system is effectively improved. In the allocation process, the strategy of foresight one step is adopted, a blocking graph model of the system obtained after foresight one step is built in real time, a corresponding loop lockingdetection and avoiding method is adopted, and the loop locking phenomenon is effectively avoided.

A game server includes: a main processing unit, connected to a user terminal via a data network, configured to perform user login; and a game processing unit configured to control the user terminal to execute a puzzle game in response to a game execution request signal from the user terminal. The game processing unit is configured to: control the user terminal to display colored blocks arranged in a matrix form of N rows and M columns in a puzzle game area, and to display one game character in an action display area; exchange a selected block with another neighboring block based on user input; determine if a matching block group of at least three consecutive blocks having the same color results from the block exchanging step; remove the matching block group; and control the game character to perform an action according to a result of the block removing step.

The invention discloses a building block automatic sorting method and system based on binocular vision, wherein the method comprises the steps of calibrating a camera; collecting the building blocks on a production line conveying belt through the calibrated camera, and generating the building block graphs; preprocessing the building block graphs to obtain the preprocessed building block graphs; carrying out image conversion on the preprocessed building block graphs to determine the colors of the building blocks; processing the images after the colors of the building blocks are determined, andobtaining the spatial position information of the building blocks; processing a picture collected by the camera, and obtaining the building block information serving as a sorting target; and sending the spatial position information and the building block information serving as the sorting target to a robot for the building block sorting. In the embodiment of the invention, a binocular vision method is used for identifying the spatial positions of the building blocks and automatically grabbing the building blocks, so that the identification and automatic grabbing of the building blocks in the industrial building block production are realized, and the problems of low efficiency and low accuracy in the building block sorting are solved.

The invention belongs to the field of computer graphics, and relates to a non-blocking graph command processing method. The method comprises the following steps that: (1) obtaining one graphic API (Application Program Interface) command from a host, and analyzing the graphic command; (2) prejudging a resource lock required by the API command, if the resource lock required by the API command is not under a locking state, issuing and executing the graphic API command, otherwise, caching the graphic API command to locality, and processing a next graphic API command; (3) carrying out graphic command function analysis on the issued graphic API command, carrying out classification processing on graphic functions, dividing the graphic functions into parameter configuration, function operations and DMA (Direct Memory Access) transportation, adopting a caching way for the parameter configuration and the function operations, and executing an initiating operation for the DAM transportation; and (4) when a host side does not have a command or a local buffer zone is full, updating the state of the resource lock. By use of the method, on the basis of guaranteeing correct functions, out-of-order execution among commands is executed and parallel execution among graphic tasks are realized, and graphic command processing efficiency is improved.

The invention discloses an automatic recognition method and system for electric secondary design cubicle and mechanism case terminal blocks. Through recognizing and distinguishing terminal block graphs of an electric secondary design cubicle and an electric secondary design mechanism case in the AUTOCAD system, marking names, serial numbers and element connecting information of the terminal blocks in the terminal block graphs of the electric secondary design cubicle and the electric secondary design mechanism case can be recognized, and then are derived into an Excel form so that a user can directly obtain the information of the electric secondary design cubicle and mechanism case terminal blocks through the Excel form, manual form drawing can be avoided, working efficiency can be improved, and the error rate of form drawing can be decreased.

The invention provides a three-dimensional scene segmentation method and device under weak supervision, and relates to the technical field of computer vision and machine learning, and the method comprises the steps of marking a point on each instance to give an instance position and a semantic category, and extending the marked point to an over-segmentation block, namely a block-level label, through an over-segmentation algorithm; building an over-segmentation block graph by taking each over-segmentation block as a node, and performing multi-level combination processing on the nodes in the over-segmentation block graph, so that labeling information is expanded to all over-segmentation blocks from labeled over-segmentation blocks, and pseudo labels are generated for all unlabeled points in a point cloud scene. Therefore, each instance in the three-dimensional scene is marked with a point to represent the position of the instance, relatively important information can be effectively marked with relatively low cost, the pseudo-label is generated for the unmarked point in the point cloud scene, and the generated pseudo-label can be directly used as supervision information to train an existing strong supervision point cloud scene segmentation network.

The invention relates to a printing bottoming method and a device adopting the same, which belong to the publishing and printing technical field. In the invention, a layout object is block-colored, the integral profile of the layout object is calculated, the block-colored layout object is contracted, a bottoming profile is corrected according to an adjacent fine line, the corrected bottoming profile and a color-blocked graph profile are combined to form a novel bottoming graph block serving a bottoming color filling area, the gradation property of the original block is reserved, and eventually, bottoming color filling is conducted to the bottoming graph block according to the type of the color block. By adopting the method and the device which are provided by the invention, the bottoming profile of the fine line can be corrected, and gradation color filling of the bottoming graph block is realized, the printing pattern quality on a transparent printing medium or a printing medium withpatterns is greatly improved.

The invention discloses an erythrocyte morphological analysis result representation method. The erythrocyte morphological analysis result representation method comprises the following steps: samples are processed to obtain morphological feature parameter values including size, shape, chromaticity and texture of each erythrocyte in the samples, at least two of the morphological feature parameter values are selected as analysis targets, and the coordinate area of an erythrocyte nine-block graph is determined according to scatter areas formed by the analysis targets of all the erythrocytes in the samples in a coordinate area; a normal erythrocyte reference range is formed by surveying the scatter areas formed by the analysis targets of erythrocytes of normal people in the erythrocyte nine-block graph; the pathological properties of abnormal erythrocytes are judged according to changes in the scatter areas formed by the analysis targets of various erythrocytes in various samples in the erythrocyte nine-block graph. According to the erythrocyte morphological analysis result representation method, an analysis result is more accurate due to the analysis of the chromaticity, size, shape and textural features of each erythrocyte in the samples, and the erythrocyte morphological analysis result representation method has higher clinic reference significance.

The invention provides a method for converting an algorithm model of a PSCAD (power system computer aided design) simulation system to a source code. The method for converting the algorithm model of the PSCAD simulation system to the source code includes following steps: building a graphic function library corresponding to a basic function block in the PSCAD simulation system; building a custom function library corresponding to a custom function block in the PSCAD simulation system; uniformly classifying the graphic function library and the custom function library which are built; converting a graphic function block to the source code. The method for converting the algorithm model of the PSCAD simulation system to the source code adopts an operating mode based on a graphical mode, achieves function block graph conversion in one to one correspondence mode and data association conversion in one to one correspondence mode, achieves the purpose that the source code is automatically generated from an algorithm model file, and can compile the source code and then download the source code into a device.

The invention discloses a self-adaptive consensus method based on main and auxiliary block graph chain structure block chain account book design, effective blocks generated by node probability are automatically divided into main blocks and auxiliary blocks, the main blocks adopt a longest chain principle to reach a consensus, a large number of auxiliary blocks are organized in a directed acyclic graph structure in the generation process of the main blocks, the transaction throughput is maximally increased, meanwhile, a block sorting algorithm is provided, and the system is prevented from recording illegal transactions by sequentially carrying out inter-graph sorting, intra-graph block sorting and intra-block transaction sorting on the graph chain structure, automatically adjusts the transaction throughput according to the network bandwidth and maximizes the bandwidth utilization rate. The method can avoid upgrading the block chain through a traditional bifurcation mode, and has the advantages of simple account book structure, efficient consensus mechanism, maximum bandwidth utilization rate, bandwidth self-adaption and the like.

The invention discloses a pedestrian target tracking method based on multi-instance learning, relates to the related field of target tracking methods, and aims to solve the problem that a target modelestablished according to a first frame in the prior art is easy to fail in tracking when the appearance of a target is greatly changed. The method comprises the following steps: S1, intercepting image data; S2, decomposing the image data into a plurality of block graphs; S3, extracting graphic features of the block graph; S4, comparing the graphic features of the block graph with the features ofthe pedestrian target; S5, constructing a classifier; S6, calculating the overlap ratio score of the examples in each classification pool of the classifier; S7, performing weight calculation on the coincidence degree score; S8, calculating and tracking a pedestrian target according to the weight, and the comparison method in the step S4 comprises the following steps: S1, stripping color features of a block graph; S2, establishing a color coincidence degree classification pool; S3, stripping the shape characteristics of the block graphs; And S4, establishing a shape coincidence degree classification pool.

The invention provides a total hip replacement preoperative planning system based on deep learning, and the system comprises a total hip joint image collection module which is used for obtaining a to-be-recognized total hip joint three-dimensional block graph; the total hip joint recognition module is used for inputting a to-be-recognized total hip joint three-dimensional block image into a trained three-dimensional segmentation neural network to obtain a thighbone area in each total hip joint two-dimensional cross section image, and the trained three-dimensional segmentation neural network is formed by a preset three-dimensional block image marked with a label of the thighbone area and a preset three-dimensional block image marked with a label of the thighbone area. The method is obtained by training a convolutional neural network. And the total hip joint three-dimensional image construction module is used for obtaining a three-dimensional image of a thighbone region according to the thighbone region in each total hip joint two-dimensional cross section image based on a three-dimensional reconstruction technology. According to the method, the total hip joint three-dimensional block graph is identified, three-dimensional modeling is carried out on the extracted thighbone area based on the three-dimensional block graph, and then the total hip joint identification precision is improved according to the thighbone area three-dimensional model.

The invention provides a logic component correlation path generation method and system. The method comprises the steps of segmenting a canvas into M*N cells in advance, and recording picture elementsin the cells by the cells; placing multiple functional block graphs used for representing logic components in the canvas, and when a logic relationship between any two functional block graphs is configured, clicking an output pin of one functional block graph and clicking an input pin of the other functional block graph; automatically calculating out and judging a type relationship between the output pin and the input pin by a background, generating an inflection point quantity of a connection line, and determining the position of each inflection point cell; and after each inflection point cell generates an inflection point picture element, enabling the cell between the two adjacent inflection point cells to generate a straight line picture element, wherein the overall connection line of all the inflection point picture elements and the straight line picture elements is namely a logic component correlation path. Through the method and the system, an engineer only needs to specify starting and terminating ends of the path; and the optimal path is automatically generated through a path algorithm.

The invention discloses a point location correction system based on communication basic resource intelligent management data, which comprises a plurality of communication function modules, and is characterized in that a geocoding module communicates with a point location determination module through a routing output module and determines communication point location information; the routing outputmodule is used for carrying out region division on the original distance data; in each area of each of the plurality of areas, the original data is converted according to a binary comparison table toobtain binary code sequences, point filling is carried out, and all the filled binary codes are merged according to the sequence of the original data to obtain original data block points, an image set is introduced, graph filling is performed on each binary code in the binary code sequence according to a set sequence, and all the graphs filled with the binary codes are merged to obtain an original data block graph.

The invention provides a machine vision communication script generation method and device based on graphical programming, and relates to the technical field of graphical programming. The method comprises the steps of: in response to dragging operation of a user on a building block graph, setting a communication graph combination on an editing interface, wherein the communication graph combination is used for describing data interaction logic for a robot to control a first visual sensor to obtain position information of a target object; in response to an input operation of a user in the communication graph combination, determining configuration data required for realizing data interaction logic; and analyzing the communication graph combination and the configuration data to obtain a script file for realizing data interaction logic. According to the machine vision communication script generation method and device based on graphical programming provided by the invention, the problem that the existing machine vision communication script file is relatively high in programming difficulty can be solved.

The invention provides a multi-thread script generation method and device based on graphical programming, and relates to the technical field of graphical programming. The multi-thread script generation method comprises the following steps of: creating an editing interface; in response to control operation of a user on the building block graphs, setting n graph combinations on the same editing interface, wherein each graph combination corresponds to one thread and comprises a first building block graph and at least one second building block graph, the first building block graph is used for identifying the graph combination to which the first building block graph belongs, the at least one second building block graph is used for indicating a code of a thread corresponding to the graph combination to which the second building block graph belongs, and n is greater than or equal to 2; and analyzing at least one second building block graph in the graph combination to which each first building block graph belongs, and generating a script file of a thread corresponding to each graph combination. According to the multi-thread script generation method based on graphical programming, the problem that an existing multi-thread script file is high in programming difficulty can be solved.

The invention provides a sliding block verification code generation method, which comprises the following steps of: obtaining a preset sliding block graph and a background graph, and generating a target graph and a plurality of interference graphs according to the sliding block graph; embedding the target graph and the interference graph into the background graph to generate a content graph; and carrying out style conversion on the content graph by utilizing a preset style image to generate a sliding block verification code, so that a user slides the sliding block graph to a position matched with the target graph to carry out verification after sending the sliding block verification code to the user side. Through style transformation, the boundary of the interference graph and the boundaryof the target graph are fused, the difficulty of identifying the boundary of the position where the sliding block graph is located by a machine is increased, and therefore the safety and reliabilityof image verification are improved.

The invention provides a mechanical arm carrying script generation method and device based on graphical programming, and relates to the technical field of graphical programming. The method comprises the steps of: in response to dragging operation of a user on building block graphs, setting a graph combination on an editing interface, wherein the graph combination comprises the building block graphs corresponding to a stacking process and/or an unstacking process, and the graph combination is used for describing the technological process of carrying materials through a mechanical arm; in response to an input operation of a user in the graph combination, determining configuration data required for realizing the technological process; and analyzing the graph combination and the configuration data to obtain a script file of the technological process. According to the mechanical arm carrying script generation method and device based on graphical programming, the problem that the script file programming difficulty of the material carrying technological process of an existing mechanical arm is large can be solved.