Abstract:
The present invention relates to an apparatus and method for generating a field line. The apparatus includes a seed point generating unit which produces a function value (f) at each point of three-dimensional volume data and generates a point list by arranging points in the large order of the produced function value, and a field line generating unit which generates the preset number of field lines of the three-dimensional volume data by using the point list generated in the seed point generating unit.
Abstract:
The present invention relates to a real time video rendering method and system using a multi-thread, and wherein the system operates a main thread and an animation thread for playing an animation video, the animation thread fetches animation data which is to be rendered for the next, processes the animation data into graphics data and stores the graphics data in a buffer while the main thread renders and displays a buffered animation graphic. Thus, the main thread swaps a main thread buffer and an animation thread buffer in the next and displays newly updated video data in a buffer on a screen. The video real time rendering method using a multi-thread according to the present invention includes: a step that a microprocessor generates a main thread and an animation thread according to a request for a video from a user; a step that the main thread renders video data which exists in a first memory and displays the video data on a screen; a step that the animation thread fetches video data which is to be rendered for the next, processes the video data into graphics data and stores the data in a second memory; a step that the main thread swaps the first memory and the second memory; and a step that the main thread renders newly updated video data in the first memory and displays the data on the screen. [Reference numerals] (110) Input unit; (120) First memory; (130) Second memory; (140) Micro processor; (150) Image processing unit
Abstract:
The present invention relates to a recording medium in which data is recorded with a data file format structure for the visualization of massive CFD parallel data and a data file format structure generation method for the same, capable of recording or storing the massive data with the data file format structure of a structure grid or a non-structure grid in the visualization by processing the massive CFD data in parallel. The present invention relates to the recording medium in which the data is recorded with the data file format structure of a grid structure for visualizing the massive CFD parallel data. The data file format structure of the structure grid comprises a meta data unit for describing feature information for the whole data including an ID and a number of data values, an ID and a number of time steps, an ID and a number of elements for data and mesh; a mesh directory unit structured with the directory of an element unit for the mesh and a directory in each time step within the element; and a data directory unit in which a directory is structured in one or more values according to each directory in the respective time steps and in which the directory of the time step within the element and the directory of an element unit for the data are structured. [Reference numerals] (110) Meta data unit; (120) Mesh directory unit; (130) Data directory unit
Abstract:
본 발명은 블록 자료구조 생성 장치, 방법, 그에 대한 기록매체 및 벡터 데이터의 필드 라인 생성 장치에 관한 것으로, 벡터 데이터를 미리 설정된 개수와 형태의 구성 블록으로 분할하는 벡터 데이터 분할부, 구성 블록이 서로 인접하는지를 판단하여 각 구성 블록의 인접 블록 정보를 추출하여 저장하고, 구성 블록의 필드 라인 생성의 우선순위 정보를 추출하여 저장하는 블록 정보 추출부, 및 인접 블록 정보와 필드 라인 생성의 우선순위 정보를 포함하는 각 구성 블록의 연결관계를 나타내는 자료구조를 생성하는 자료구조 생성부를 포함한다.
Abstract:
PURPOSE: A method for generating and terminating an MPI(Massage Passing Interface) parallel process, a device thereof, and a recording medium thereof are provided to dynamically generate and assign the MPI parallel process by considering a use situation of a current system, thereby efficiently enabling the using of computing resources. CONSTITUTION: A parallel process management unit(100) is located in a host node among computing nodes which process an MPI parallel process and receives system resource use information of the computing nodes. The parallel process management unit arranges the computing nodes corresponding to order which the amount of system resource use is small and assigns the computing nodes in order that a parallel process is generated corresponding to the order. A parallel process generation unit(200) generates a parallel process in an assigned computing node. [Reference numerals] (100,AA,BB,CC) MPI parallel process management unit; (200) MPI parallel process generation unit
Abstract:
PURPOSE: Application independent visualization protocol service system and a method for distributed parallel processing of mass data are provided to perform visualization of parallel distribution. CONSTITUTION: Application independent visualization protocol service system(100) includes a GIP(Glove Interface Protocol) client(110), a communication network(120) and a GIP server(130). The GIP client transmits a GIP request message inputted from a user to the GIP server through a GIP packet according to the GIP protocol and receives a GIP response message from the GIP server. The GIP server supports synchronous and asynchronous protocols for the GIP protocol, generates a GIP response message about a visualization result and transmits the GIP response message to the GIP client through a GIP packet according to the GIP protocol if the GIP server receives a GIP request message from the GIP client. [Reference numerals] (110) GIP client; (120) Communication network; (130) GIP server
Abstract:
PURPOSE: A transform function creation apparatus and method thereof are provided to minimize the loss of information in low resolution for numeric data by creating transform functions through non-linear histogram values. CONSTITUTION: A linear histogram creation module(110) creates a linear histogram. A non-linear histogram creation module(130) creates equalized non-linear histogram by distributing range values of the linear histogram. A transform function creation module(150) samples feature values in the non-linear histogram. After interpolation functions of each feature value is determined, the transform function creation module calculates non-linear histogram values for each feature values. The transform function creation module creates transform functions.