公开(公告)号：EP0525592A2

公开(公告)日：1993-02-03

申请号：EP92112449.1

申请日：1992-07-21

申请人： HITACHI, LTD. ,  HITACHI INFORMATION & CONTROL SYSTEMS, INC.

发明人： Hebishima, Shingo, Daini Satsuki Haitsu 208 ,  Matsuba, Ikuo

IPC分类号： G06F15/80

CPC分类号： G06N3/0418

摘要： An information processing system includes a data modifying module (102) for modifying input data to the system, a fractal dimension calculating module (103) for determining a fractal dimension of the input data, a main processing decision module (104) and a main processing module (105). By checking the fractal dimension of the data inputted to the main processing module (105), decision is made as to whether or not the main processing is required. If so, the main processing module (105) is activated to perform adjustment of parameters, modification of output or the like on the basis of the fractal dimension. By introducing as an index the fractal dimension of the data inputted to the system, it is possible to condition previously the system so that most appropriate effect can be attained by utilizing the input data.

摘要翻译： 一种信息处理系统，包括用于将输入数据修改到系统的数据修改模块（102），用于确定输入数据的分形维数的分形维数计算模块（103），主处理决策模块（104）和主处理 模块（105）。 通过检查输入到主处理模块（105）的数据的分形维数，决定是否需要主处理。 如果是，则主处理模块（105）被激活以基于分形维数来执行参数的调整，输出的修改等。 通过引入输入到系统的数据的分形维数作为索引，可以先前对系统进行调整，以便通过利用输入数据可以获得最适当的效果。

公开(公告)号：EP1115070A1

公开(公告)日：2001-07-11

申请号：EP00946334.0

申请日：2000-07-14

申请人： Sony Corporation

发明人： UGAJIN, Ryuichi ,  KUROKI, Yoshihiko ,  ISHIBASHI, Akira ,  HIRATA, Shintaro

IPC分类号： G06F17/10

CPC分类号： G06N3/0418

摘要： Fractal structures are grown from a plurality of start sites. The fractal structures growing from their respective start sites progress their growth while interacting with each other to form fractal structures coupled to each other and create a neural network. Growth rate from a specific start site is determined by the probability that a material reaches a portion already grown from a remote site in a diffusion process and the probability that a growth promotion factor reaches the portion already grown from portions grown from the other start sites in the diffusion process. Anisotropy, if necessary, may be introduced into the space where the fractal structures are grown.

摘要翻译： 分形结构从多个起始位点生长。 从它们各自的起始位点生长的分形结构在彼此相互作用的同时进行其生长，以形成彼此耦合的分形结构并创建神经网络。 来自特定起始位点的生长速率由材料到达在扩散过程中从远端位点生长的部分的概率以及生长促进因子到达已经从其它起始位点生长的部分生长的部分的概率来确定 扩散过程。 如果需要，可以将各向异性引入分形结构生长的空间中。

公开(公告)号：EP0425315A2

公开(公告)日：1991-05-02

申请号：EP90311787.7

申请日：1990-10-26

申请人： THE GENERAL ELECTRIC COMPANY, p.l.c.

发明人： Stark, Jaroslav

IPC分类号： G06F15/72

CPC分类号： G06N3/0635 ,  G06N3/0418

摘要： In a process for generating images using the mathematical framework of fractals implemented as a neural network architecture, processing for each pixel of the image is carried out by a respective processing element (10,11). All of the elements operate in parallel, thereby achieving very fast processing. The generated image is determined by the strengths of the connections between the processing elements. The processing elements may be individual elements or may be implemented in a computer. The image to be generated is preferably encoded as an Iterated Function System.

摘要翻译： 在使用实现为神经网络架构的分形数学框架生成图像的过程中，通过相应的处理元件（10,11）执行图像的每个像素的处理。 所有的元件都是并行操作的，从而实现了非常快速的处理。 生成的图像由处理元件之间的连接的强度确定。 处理元件可以是单独元件或者可以在计算机中实现。 要生成的图像优选地被编码为迭代函数系统。

公开(公告)号：EP0525592A3

公开(公告)日：1994-06-08

申请号：EP92112449.1

申请日：1992-07-21

申请人： HITACHI, LTD. ,  HITACHI INFORMATION & CONTROL SYSTEMS, INC.

发明人： Hebishima, Shingo, Daini Satsuki Haitsu 208 ,  Matsuba, Ikuo

IPC分类号： G06F15/80

CPC分类号： G06N3/0418

摘要： An information processing system includes a data modifying module (102) for modifying input data to the system, a fractal dimension calculating module (103) for determining a fractal dimension of the input data, a main processing decision module (104) and a main processing module (105). By checking the fractal dimension of the data inputted to the main processing module (105), decision is made as to whether or not the main processing is required. If so, the main processing module (105) is activated to perform adjustment of parameters, modification of output or the like on the basis of the fractal dimension. By introducing as an index the fractal dimension of the data inputted to the system, it is possible to condition previously the system so that most appropriate effect can be attained by utilizing the input data.

公开(公告)号：EP0425315A3

公开(公告)日：1992-10-07

申请号：EP90311787.7

申请日：1990-10-26

申请人： THE GENERAL ELECTRIC COMPANY, p.l.c.

发明人： Stark, Jaroslav

IPC分类号： G06F15/72

CPC分类号： G06N3/0635 ,  G06N3/0418

摘要： In a process for generating images using the mathematical framework of fractals implemented as a neural network architecture, processing for each pixel of the image is carried out by a respective processing element (10,11). All of the elements operate in parallel, thereby achieving very fast processing. The generated image is determined by the strengths of the connections between the processing elements. The processing elements may be individual elements or may be implemented in a computer. The image to be generated is preferably encoded as an Iterated Function System.

摘要翻译： 在使用实现为神经网络架构的分形数学框架生成图像的过程中，通过相应的处理元件（10,11）执行图像的每个像素的处理。 所有的元件都是并行操作的，从而实现了非常快速的处理。 生成的图像由处理元件之间的连接的强度确定。 处理元件可以是单独元件或者可以在计算机中实现。 要生成的图像优选地被编码为迭代函数系统。