Matches in SemOpenAlex for { <https://semopenalex.org/work/W841941974> ?p ?o ?g. }
Showing items 1 to 78 of
78
with 100 items per page.
- W841941974 endingPage "652" @default.
- W841941974 startingPage "643" @default.
- W841941974 abstract "본 연구에서는 신경망 기반 독립성분분석의 분리성능을 개선하기 위해 할선법과 모멘트의 조합형 고정점 알고리즘을 제안하였다. 할선법은 독립성분 상호간의 정보를 최소화하는 목적함수의 근을 근사적으로 구함으로써 계산과정을 단순화하여 좀 더 개선된 분리성능을 얻기 위함이고, 모멘트는 계산과정에서 발생하는 발진을 억제하여 보다 빠른 분리속도를 얻기 위함이다. 이렇게 하면 할선법이 가지는 근사성에 따른 우수성과 과거의 속성을 반영하여 발진을 억제하는 모멘트의 우수성을 동시에 살릴 수 있다. 제안된 알고리즘을 <TEX>$256times{256}$</TEX> 픽셀의 8개 지문과 <TEX>$512times{512}$</TEX> 픽셀의 10개 영상으로부터 임의의 혼합행렬에 따라 생성된 복합지문과 복합영상을 각각 대상으로 시뮬레이션 한 결과, 뉴우턴법에 기초한 기존의 알고리즘과 할선법만에 기초한 알고리즘보다 각각 우수한 분리률과 빠른 분리속도가 있음을 확인하였다. 또한 할선법의 이용은 뉴우턴법을 이용한 고정점 알고리즘보다 초기값에도 덜 의존하며, 문제의 규모가 커짐에 따른 비현실적인 분리시간도 해결할 수 있음을 확인하였다. This paper proposes an efficient hybrid fixed-point (FP) algorithm for improving performances of the independent component analysis (ICA) based on neural networks. The proposed algorithm is the FP algorithm based on secant method and momentum for ICA. Secant method is applied to improve the separation performance by simplifying the computation process for estimating the root of objective function, which is to minimize the mutual informations of the independent components. The momentum is applied for high-speed convergence by restraining the oscillation if the process of converging to the optimal solution. It can simultaneously achieve a superior properties of the secant method and the momentum. The proposed algorithm has been applied to the composite fingerprints and the images generated by random mixing matrix in the 8 fingerprints of <TEX>$256times{256}$</TEX>-pixel and the 10 images of <TEX>$512times{512}$</TEX>-pixel, respectively. The simulation results show that the proposed algorithm has better performances of the separation speed and rate than those using the FP algorithm based on Newton and secant method. Especially, the secant FP algorithm can be solved the separating performances depending on initial points settings and the nonrealistic learning time for separating the large size images by using the Newton FP algorithm." @default.
- W841941974 created "2016-06-24" @default.
- W841941974 date "2002-10-01" @default.
- W841941974 modified "2023-09-25" @default.
- W841941974 title "Independent Component Analysis Based on Neural Networks Using Hybrid Fixed-Point Algorithm" @default.
- W841941974 cites W1984121966 @default.
- W841941974 cites W2019502123 @default.
- W841941974 cites W2099741732 @default.
- W841941974 doi "https://doi.org/10.3745/kipstb.2002.9b.5.643" @default.
- W841941974 hasPublicationYear "2002" @default.
- W841941974 type Work @default.
- W841941974 sameAs 841941974 @default.
- W841941974 citedByCount "0" @default.
- W841941974 crossrefType "journal-article" @default.
- W841941974 hasBestOaLocation W8419419741 @default.
- W841941974 hasConcept C111919701 @default.
- W841941974 hasConcept C11413529 @default.
- W841941974 hasConcept C121332964 @default.
- W841941974 hasConcept C127162648 @default.
- W841941974 hasConcept C154945302 @default.
- W841941974 hasConcept C158622935 @default.
- W841941974 hasConcept C160633673 @default.
- W841941974 hasConcept C162324750 @default.
- W841941974 hasConcept C2777303404 @default.
- W841941974 hasConcept C31258907 @default.
- W841941974 hasConcept C33923547 @default.
- W841941974 hasConcept C41008148 @default.
- W841941974 hasConcept C45374587 @default.
- W841941974 hasConcept C50522688 @default.
- W841941974 hasConcept C50644808 @default.
- W841941974 hasConcept C51432778 @default.
- W841941974 hasConcept C57869625 @default.
- W841941974 hasConcept C62520636 @default.
- W841941974 hasConcept C68339613 @default.
- W841941974 hasConcept C85189116 @default.
- W841941974 hasConcept C93926214 @default.
- W841941974 hasConceptScore W841941974C111919701 @default.
- W841941974 hasConceptScore W841941974C11413529 @default.
- W841941974 hasConceptScore W841941974C121332964 @default.
- W841941974 hasConceptScore W841941974C127162648 @default.
- W841941974 hasConceptScore W841941974C154945302 @default.
- W841941974 hasConceptScore W841941974C158622935 @default.
- W841941974 hasConceptScore W841941974C160633673 @default.
- W841941974 hasConceptScore W841941974C162324750 @default.
- W841941974 hasConceptScore W841941974C2777303404 @default.
- W841941974 hasConceptScore W841941974C31258907 @default.
- W841941974 hasConceptScore W841941974C33923547 @default.
- W841941974 hasConceptScore W841941974C41008148 @default.
- W841941974 hasConceptScore W841941974C45374587 @default.
- W841941974 hasConceptScore W841941974C50522688 @default.
- W841941974 hasConceptScore W841941974C50644808 @default.
- W841941974 hasConceptScore W841941974C51432778 @default.
- W841941974 hasConceptScore W841941974C57869625 @default.
- W841941974 hasConceptScore W841941974C62520636 @default.
- W841941974 hasConceptScore W841941974C68339613 @default.
- W841941974 hasConceptScore W841941974C85189116 @default.
- W841941974 hasConceptScore W841941974C93926214 @default.
- W841941974 hasIssue "5" @default.
- W841941974 hasLocation W8419419741 @default.
- W841941974 hasOpenAccess W841941974 @default.
- W841941974 hasPrimaryLocation W8419419741 @default.
- W841941974 hasRelatedWork W1530047662 @default.
- W841941974 hasRelatedWork W1586585988 @default.
- W841941974 hasRelatedWork W2128767608 @default.
- W841941974 hasRelatedWork W2134719536 @default.
- W841941974 hasRelatedWork W2373276648 @default.
- W841941974 hasRelatedWork W2379932168 @default.
- W841941974 hasRelatedWork W3001734434 @default.
- W841941974 hasRelatedWork W4310449831 @default.
- W841941974 hasRelatedWork W657818131 @default.
- W841941974 hasRelatedWork W841941974 @default.
- W841941974 hasVolume "9B" @default.
- W841941974 isParatext "false" @default.
- W841941974 isRetracted "false" @default.
- W841941974 magId "841941974" @default.
- W841941974 workType "article" @default.