@article{oai:bunkyo.repo.nii.ac.jp:00003466,
 author = {鈴木, 昇一},
 journal = {情報研究, Information and Communication Studies},
 month = {1996-01-01, 2012-01-17},
 note = {多段階認識法で用いられる2種類のモデル構成作用素がRBF法、wavelet展開法を適用して、構成されている。得られたモデル構成作用素T.の像Tφは、原パターンφのパターンモデルといわれ、RBF法、wave-let展開法で得られているパターンが正のスカラー定数倍についての不変性、ベキ等性を備えている樣に構成し直されたものである。認識システムは原パターンφを恰も、Tφかのごとく、錯覚して、φの認識処理をすることが可能になる。
\n　The two applications of the method of RBF and the theory of wavelets will help us to construct new models of the corresponding pattern. Two images of two kinds of model-construction operators T obtained here as their applications are called two pattern-models. Two pattern-models remains invariant under multiplication of any positive scalar and are possessed of idempotency. Two models are needed to design a faculties of multi-stage recognition. A recognition system can extract features from Tφ instead of φ and therefore can classify Tφ exactly as though Tφ were φ.},
 pages = {71--131},
 title = {Radial-Basis Function Networks, Wavelet-Based Networks を用いたモデル構成作用素の構成法},
 volume = {17},
 year = {}
}