Depth Of Element Transfer And Separation-identification Of Implicit Image

Rough sets theory, as a new mathematical approach to the expression, study and induction for imprecise, uncertain or incomplete information or knowledge, has become one of the research focuses in artificial intelligence, information sciences and so on, for it has been widely and successfully applied to machine learning, knowledge discovery, data mining, decision support and systems analysis and so on. Materialistic dialectic tells us that things are constantly changing and developping, the static and unchanging world does not exist. In system science, the system (such as financial system, risk investment system, medical diagnosis system, etc) problems are often dynamic. The classical Pawlak rough set is just static theory and helpless for dealing with dynamic problems. In 2005, S-rough sets which has dynamic characteristic was proposed by Prof. Shi Kaiquan in Shandong University by improving Pawlak rough sets, S-rough set expands space of the Pawlak rough set and provide a theoretical support for the research of dynamic data mining, dynamic knowledge discovery and system dynamic rough characteristic, etc. With the in-depth study of S-rough set, Prof. Shi discovered that it was difficult for S-rough sets to deal with the system law problems, in 2005, he brought the analysis tool of function into it and proposed function S-rough sets which has dynamic law characteristics. Function S-rough sets is a theoretical approach to system law-mining, system fault detection-recognition, invest system warning ayalysis, profit estimate, image delivery, material discovery, biomedical engineering, etc.Based on the summarization of predecessor? works, this dissertation presents the primary researches as follows:1. Element transfer is the basic and important notions in S-rough sets theory, for the reason why S-rough theory was proposed is that the classical rough set lacks for dynamic characteristic, the direct result of introduction of dynamic characteristic is element transfer. Element transfer contains element transfer-in and transfer-out, which have different depths. How to describe the depth of element transfer is very important to understand the connotation of element transfer. This paper defines the concepts of interior point, exterior point, interior accumulation point, exterior accumulation point of element transfer, etc, and subsequently extension coefficient function and rough membership degree function and so on are given. Depended on those concepts above, ordered relation of element transfer eigenvector and its comparision method are put forward, in the end, the application in information recognition of stock market system is given.2. Based on function S-rough sets, the concepts of implict image, rough law, attribute distance, state coefficient and so on are defined, after analyzing signification of those concepts in the image identification system and risk investment system, image identification criterion, attribute control theorem, attribute control criterion and their application are given.