NCHU Institutional Repository CRIShttp://ir.lib.nchu.edu.twDSpace 數位典藏系統是用來獲取、儲存、索引、散佈數位研究資料。Tue, 26 Mar 2019 21:22:54 GMT2019-03-26T21:22:54Z50491結實M-碼與分離型域http://hdl.handle.net/11455/18241標題: 結實M-碼與分離型域; Solid m-codes and disjunctive domains
作者: 喻石生; YU, SHI-SHENG
摘要: 令x 是一個大有限的字母集且令X* 是X 所生成的自由域(free monoid)。一個言語D
X* ,對若對任一x,y D,(X Y(RA)X (RA)，則A 是一個（若）分離型言語，其中
A X*,此時稱D 是一個（右）分離域。一種特別的有限碼，稱之為結實m 一碼，要被
提出來。本論文的目的是研究結實m 一碼的一些性質，及將分離域及右分離域特徵化
。在本論文的第二部份中，我們證明了下列定理：
令A=(S,X,M,q0,F) 是一個有限自動機且令|S|=n≧2 則，
(A)T(A)∩Q≠ 若且唯若A 接受一個質字其長度小或等於3(n-1)。
(B)T(A)∩Q|=∞若且唯若A 接受一個質字而
n≦1g(y)≦3(n-1)
此外，還證明了|T(A)∩Q|∞ 時自動機A 的一些性質。如此我們可以很清晰地來判斷
自動機與集合Q 的關係；換句話說，由自動機的觀點，我們可以清楚地判斷質字存在
一個正規言語中的狀況。
1.Hopcroft, J.E. and Ullman, J.D.,"Formal Languages and their Relation to
Automata," Addison-Wesley, Massachusetts, 1976.
2.Ito Masami, Shy, H.J. and Katsura M. Regular Domains.(Submitted for publi
cation).
3.Lallement, G. "Semigroups and Combinatorial Applications,"John Wiley and
Sons, New York(1979).
4.Lothaire. M.,Combinatorics on Words, Addision-Wesley, Reading,(1983) .
5.Shyr,H.J.,Disjunctive Languages on a Free Monoid, Information and Contro
l, Vol. 34,(1977)123-129.
6.Shyr,H.J.,"Free Monoids and Languages,"Lecture Notes, Department of Math
ematics, Soochow Unviersity, Taipei, Taiwain,(1979).
7.Shyr, H.J., Order Catenation and Regular Free Disjunctive Languages, Inf
ormation and Control, Vol. 46, No. 3, (1980)257-269.
8.Shyr, H.J., and Tseng, D.C., Some Properties of Dense Languages, Soochow
Journal of Mathematics, Vol. 10,(1984)127-131.
Wed, 01 Jan 1986 00:00:00 GMThttp://hdl.handle.net/11455/182411986-01-01T00:00:00Zd-minimal languageshttp://hdl.handle.net/11455/37961標題: d-minimal languages
作者: Yu, S.S.; 喻石生
摘要: A word v is said to be a proper d-factor of a word u if v not equal u and u = vx = yv for some words x, y. The family of words which have i distinct proper d-factors is denoted by D(i). According to the number of distinct proper d-factors of words, the free semigroup X+ generated by X can be expressed as the disjoint union of D(i)'s. Words in D(1) are called d-minimal words. d-Minimal words are often called non-overlapping words, dipolar words or unborded words. In this paper, we study the relationship between D-i(I) and D(i) concerning the basic properties of decompositions and catenations of words. We give characterizations of words in D-2(1)boolean AND D(I) and D(2). We also show that sets D-i(I)\D(j) and D(j)\D-i(1) are disjunctive. It is known that every disjunctive language is dense but not regular. We obtain the results that X+D(1) and X+D(2) are regular but X+D(i) is disjunctive for every i greater than or equal to 4. Served as an example of disjunctive d-minimal context-free languages, a disjunctive d-minimal context-free language is constructed. Moreover, we show that the well-known Dyck language is a free semigroup generated by a d-minimal bifix code. The languages of which the catenations consist of d-minimal words are studied in this paper too. That is, some properties of d-minimality-annihilators of languages are investigated. (C) 1998 Elsevier Science B.V. All rights reserved.
Thu, 01 Jan 1998 00:00:00 GMThttp://hdl.handle.net/11455/379611998-01-01T00:00:00ZBi-catenation and shuffle product of languageshttp://hdl.handle.net/11455/37954標題: Bi-catenation and shuffle product of languages
作者: Shyr, H.J.; 喻石生; Yu, S.S.
摘要: The shuffle product of two words consists of all words obtained by inserting one word into another word sparsely. The shuffle product of two languages is the union of all the shuffle products of two words taken one from each of these two languages. The bi-catenation of two languages A and B is the set A * B = AB boolean OR BA. A non-empty word which is not a power of any other word is called a primitive word. A language is a prefix code if no word in this language is a prefix of any other word in this language. This paper is devoted to the investigation of the elementary properties of bi-catenation and shuffle product of languages. The families of prefix codes, disjunctive languages and languages consisting of primitive words with respective to these two operations are studied. We characterize languages of which the bi-catenation or the shuffle product with any non-empty word are prefix codes. We also derive that for any bifix code A, both A * Q and A * Q((n)), n greater than or equal to 2, are disjunctive languages, where Q is the set of all primitive words over an alphabet X with more than one letter and Q((n)) = {f(n) \f is an element of Q}. For the shuffle product case, surprisingly a lozenge Q is a regular language, where a is a letter of the alphabet X.
Thu, 01 Jan 1998 00:00:00 GMThttp://hdl.handle.net/11455/379541998-01-01T00:00:00Zd-Words and d-languageshttp://hdl.handle.net/11455/37956標題: d-Words and d-languages
作者: Fan, C.M.; 喻石生; Shyr, H.J.; Yu, S.S.
摘要: Let X be a finite alphabet containing more than one letter. A d-primitive word u over X is a non-overlapping word in the sense that no proper prefix of u is a suffix of u. D(1) is the set of all d-primitive words over X and D is the set of all positive powers of all words in D(1). Every language in D will be called a d-language. In this paper, we study some algebraic properties of d-primitive words and d-languages relative to formal language theory and codes. We,show that there are infinitely many cyclic-square-free words over alphabet with three letters. A characterization of three elements codes in D(1) is obtained and we prove that every regular component in D(1) is either a prefix code or a suffix code.
Thu, 01 Jan 1998 00:00:00 GMThttp://hdl.handle.net/11455/379561998-01-01T00:00:00ZReversible fragile watermarking scheme for three-dimensional modelshttp://hdl.handle.net/11455/38015標題: Reversible fragile watermarking scheme for three-dimensional models
作者: Wang, J.T.; 喻石生; Wang, P.C.; Yu, S.S.
摘要: Watermarking is a very important technique for protecting the authorization of digital content, such as still images, video streams, audio streams, and 3-D models. Most fragile watermarking schemes for 3-D models are not reversible. Published reversible or irreversible fragile watermarking schemes for 3-D models have many drawbacks, such as the causality problem, not being blind, or being unable to locate changed regions. To overcome these drawbacks, a reversible fragile watermarking scheme for 3-D models in spatial domain is proposed. Principal component analysis (PCA) is employed to produce the PCA coordinate system and make the system robust against similarity transformation attacks. PCA, together with a novel interval embedding technique, provide the blind reversibility of the proposed scheme. Intervals larger than a threshold and smaller than another threshold are adopted as legitimate to avoid large distortion. Moreover, experimental results show that the proposed scheme can overcome the causality, convergence, and confusion problems; provide both tampering detection and better embedding rate and requiring a small key size. (C) 2009 Society of Photo-Optical Instrumentation Engineers. [DOI: 10.1117/1.3223635]
Thu, 01 Jan 2009 00:00:00 GMThttp://hdl.handle.net/11455/380152009-01-01T00:00:00ZDependences related to strict binary relationshttp://hdl.handle.net/11455/38019標題: Dependences related to strict binary relations
作者: Hsiao, H.K.; 喻石生; Yeh, Y.T.; Yu, S.S.
摘要: In this paper we study dependences on a free monoid X* and their relations to strict binary relations on X*. As a consequence we characterize strict binary relations by the notion of dependence. Moreover, some results concerning the relationship between dependences and codes are also obtained. A visualization of this work is provided by the theory of graph. (c) 2005 Elsevier B.V. All rights reserved.
Sat, 01 Jan 2005 00:00:00 GMThttp://hdl.handle.net/11455/380192005-01-01T00:00:00ZPost-plus languageshttp://hdl.handle.net/11455/38021標題: Post-plus languages
作者: Yu, S.S.; 喻石生
摘要: In this paper, a post-plus language is defined as a language which contains all the catenations of each word in the language with a proper suffix of this word. The set of all d-primitive words is a natural post-plus language. The family of all post-plus languages is a subfamily of all non-counting languages. Some basic properties and characterizations of post-plus languages are investigated. We obtain that a post-plus language spanned by a word over an alphabet with two letters is context-free if and only if the language is regular. Some general properties of post-plus languages related to code, dense property and formal language theory such as the nature of context-free, context-sensitive languages are also studied in this paper. (C) 2001 Elsevier Science B.V. All rights reserved.
Mon, 01 Jan 2001 00:00:00 GMThttp://hdl.handle.net/11455/380212001-01-01T00:00:00ZSquare-free-preserving and primitive-preserving homomorphismshttp://hdl.handle.net/11455/37959標題: Square-free-preserving and primitive-preserving homomorphisms
作者: Hsiao, H.K.; 喻石生; Yeh, Y.T.; Yu, S.S.
摘要: The aim of this paper is to investigate homomorphisms which preserve square-free languages or primitive languages. A characterization of square-free-preserving homomorphisms is presented. We show that every square-free-preserving homomorphism is primitive-preserving. Strongly cube-free-preserving homomorphisms axe also studied.
Wed, 01 Jan 2003 00:00:00 GMThttp://hdl.handle.net/11455/379592003-01-01T00:00:00ZThe decodability and correctability of codeshttp://hdl.handle.net/11455/37990標題: The decodability and correctability of codes
作者: Hsiao, H.K.; 喻石生; Lin, H.H.; Yu, S.S.
摘要: Given a noisy communication channel gamma, the aim of this paper is to investigate gamma-decodable codes and gamma-correctable codes. Examples are given for channels with possible errors caused by deletions. The gamma-correctability does not imply gamma-decodability but ensures error recovery. In general, the properties of gamma-correctability and gamma-decodability have little in common. Some general properties concerning gamma-decodable codes and gamma-correctable codes are derived. Moreover, we propose a special kind of codes called gamma-separable codes having a unique decodability.
Fri, 01 Jan 2010 00:00:00 GMThttp://hdl.handle.net/11455/379902010-01-01T00:00:00ZA characterization of local regular languageshttp://hdl.handle.net/11455/37967標題: A characterization of local regular languages
作者: Yu, S.S.; 喻石生
摘要: As every non-empty word is a power of a unique primitive word, a set of primitive roots of a language is like an independent subset of a vector space. A language having finitely many primitive roots is called a local language. The purpose of this paper is to characterize local regular languages. We show that whether a regular language is local or not is decidable. In the meanwhile, two characterizations of local regular languages are derived. (C) 1998 Elsevier Science B.V. All rights reserved.
Thu, 01 Jan 1998 00:00:00 GMThttp://hdl.handle.net/11455/379671998-01-01T00:00:00ZAlpha-gamma equalization-enhanced hand radiographic image segmentation schemehttp://hdl.handle.net/11455/38013標題: Alpha-gamma equalization-enhanced hand radiographic image segmentation scheme
作者: Lin, H.H.; 喻石生; Shu, S.G.; Kuo, S.W.; Wang, C.H.; Chan, Y.P.; Yu, S.S.
摘要: Bone age assessment of children is a procedure frequently performed in pediatric radiology. The feature extractions of metaphyseal and epiphyseal regions are crucial to automatic bone age assessment. The first step of feature extraction is applying a segmentation scheme to find exact regions of epiphysis and metaphysis. A segmentation method is normally based on both intensity information and the relative location of pixels. There is a fundamental problem; when the intensity contrast of soft tissue and bony tissue is poor, bony and soft tissue cannot easily be separated. We propose an alpha-gamma equalization method to increase the intensity contrast between bony and soft tissue. Sobel, two-means, Canny edge-detection, and watershed methods are applied to illustrate the effect of this method on initial segmentation. Adaptive two-means and gradient vector flow snake are adopted for the final segmentation. Experimental results reveal that alpha-gamma equalization-enhanced two-means initial segmentation with an adaptive two-means clustering scheme can greatly reduce the average error measurements of segmentations. These are evaluated according to the following measurements: misclassification error, edge mismatch, region nonuniformity, relative foreground area eror, and modified Hausdorff distance. Furthermore, the experimental results show that the proposed scheme provides increased stable performance for the segmentation of epiphyseal/metaphyseal regions. (C) 2009 Society of Photo-Optical Instrumentation Engineers. [DOI: 10.1117/1.3242852]
Thu, 01 Jan 2009 00:00:00 GMThttp://hdl.handle.net/11455/380132009-01-01T00:00:00ZTHE EFFECT OF ROI NORMALIZATION FOR HAND RADIOGRAPHIC IMAGE SEGMENTATIONhttp://hdl.handle.net/11455/38009標題: THE EFFECT OF ROI NORMALIZATION FOR HAND RADIOGRAPHIC IMAGE SEGMENTATION
作者: Lin, H.H.; 喻石生; Chiang, W.C.; Shu, S.G.; Shih, L.M.; Yu, S.S.
摘要: The accuracy of epiphyseal/metaphyseal segmented results of a clustering scheme and the consistency of feature measurement depend on a reliable extraction of regions of interest (ROI) in an automatic bone age assessment system. ROI normalization is an important step for these purposes. The related modeling techniques are different according to the characteristics and applications of the processed images. Most of the epiphyseal metaphyseal ROI extraction schemes in the literature either only use hand normalization or do not use any normalization processing. Here, a novel ROI normalization scheme is proposed, which includes hand rotation to standard orientation as well as the individual finger rotation. An evaluation of the effect of the ROI normalization scheme in experiments derives from our distance approach adaptive 2-means clustering method and literature reviewed methods with regard to the final epiphyseal metaphyseal segmentation. Experimental results reveal that the proposed ROI normalization scheme provides a very well automated segmentation ability to separate accurately the epiphysis and metaphysis from the soft tissue of hand radiographs at the early stage of skeletal development. Furthermore, the experimental results show that the proposed approach provides a more stable performance for the segmentation of epiphyseal/metaphyseal regions.
Sat, 01 Jan 2011 00:00:00 GMThttp://hdl.handle.net/11455/380092011-01-01T00:00:00ZEXCLUDING BACKGROUND INITIAL SEGMENTATION FOR RADIOGRAPHIC IMAGE SEGMENTATIONhttp://hdl.handle.net/11455/38006標題: EXCLUDING BACKGROUND INITIAL SEGMENTATION FOR RADIOGRAPHIC IMAGE SEGMENTATION
作者: Shu, S.G.; 喻石生; Lin, H.H.; Kuo, S.W.; Yu, S.S.
摘要: Bony tissue extraction of phalangeal ROI of radiographic images highly affects the results of feature extraction and bone age assessment. The performance of most segmentation techniques, such as k-means, snakes, and so oil, relies on the precision of a given initial segmentation. Most papers provide the initial segmentation by random or manual choices. Here, an excluding background initial segmentation method is proposed to overcome the initial segmentation problem for feature extraction oil phalangeal ROI. Sobel, 2-means, Canny edge-detection and watershed methods are used to provide an initial segmentation and compared to the proposed method by applying adaptive 2-means and GVF snake to do the finial segmentation. The experiment results show that the proposed excluding background initial segmentation method together with adaptive 2-means clustering method provides a very well automatic segmentation ability to separate accurately the epiphysis and metaphysis from the soft tissue of hand radiographs at the early stage of skeletal development.
Thu, 01 Jan 2009 00:00:00 GMThttp://hdl.handle.net/11455/380062009-01-01T00:00:00ZThe disjunctivities of omega-languageshttp://hdl.handle.net/11455/37964標題: The disjunctivities of omega-languages
作者: Yeh, Y.T.; 喻石生; Yu, S.S.
摘要: An omega-language over a finite alphabet X is a set of infinite sequences of letters of X. In this paper, we consider a congruence I-L on X* and a congruence O-L on X-omega introduced by an omega-language L, called the infinitary syntactic congruence and the omega-syntactic congruence of L, respectively. If I-L is the equality, then L is called an I-disjunctive omega-language. Some results concerning I-disjunctive omega-languages are obtained. If OL is the equality, then L is called an O-disjunctive omega-language. A construction of O-disjunctive omega-languages is given. Properties concerning O-disjunctive omega-languages are investigated. An omega-language such that every omega-dense subset of it is an O-disjunctive omega-language is called a completely O-disjunctive omega-language. A completely O-disjunctive omega-language is given. Properties concerning completely O-disjunctive omega-languages are studied in this note. (C) 2003 Elsevier Science B.V. All rights reserved.
Wed, 01 Jan 2003 00:00:00 GMThttp://hdl.handle.net/11455/379642003-01-01T00:00:00ZAdaptive DE-based reversible steganographic technique using bilinear interpolation and simplified location maphttp://hdl.handle.net/11455/38011標題: Adaptive DE-based reversible steganographic technique using bilinear interpolation and simplified location map
作者: Liu, Y.C.; 喻石生; Wu, H.C.; Yu, S.S.
摘要: In this paper, an adaptive DE-based reversible steganographic scheme with bilinear interpolation and simplified location map is proposed. In traditional reversible difference expansion (DE) scheme, it suffers from two problems: the embeddable location is considered insufficient and the embedding payload control capability in single layer embedding is weak. For the first problem, the kernel of bilinear interpolation is applied to effectively improve the number of the embeddable location while the quality of the stego-image can be maintained at a good level. In addition, the proposed simplified location map is used for the existing adaptive embedding rule to improve the second problem where the secret data can be adaptively embedded and also the load of additional information can be reduced. The experimental results revealed that the proposed scheme presented better visual quality of the stego-image and carried larger embedding payload than some other revised DE schemes, such as Alattar's and Lee's schemes.
Sat, 01 Jan 2011 00:00:00 GMThttp://hdl.handle.net/11455/380112011-01-01T00:00:00Z非消去言語的研究http://hdl.handle.net/11455/49905標題: 非消去言語的研究; A Study of Non-Cancellable Languages
作者: 喻石生
Mon, 01 Jan 2001 00:00:00 GMThttp://hdl.handle.net/11455/499052001-01-01T00:00:00ZA protozoan parasite extraction scheme for digital microscopic imageshttp://hdl.handle.net/11455/38127標題: A protozoan parasite extraction scheme for digital microscopic images
作者: Lai, C.H.; 蔡孟勳; Yu, S.S.; Tseng, H.Y.; Tsai, M.H.; 喻石生
摘要: Pathogenic protozoan parasites can cause human to get many diseases, such as, amoebiasis, typhoid fever and cholera, etc. Different protozoan parasites vary greatly in their structural and biochemical properties. Digital images are extensively applied to medical fields for doctors and pathologists to analyze pathological sections and further diagnose diseases. The aim of this paper is to develop protozoan parasite extraction techniques to segment protozoan parasites from microscopic images. The proposed scheme has precise segmentation ability even if the image is with poor quality or complex background. Experimental results show that the proposed scheme can gain 96.64% average correct rate, and about 0.04, 0.45 and 0.06 of the average error rates: misclassification error (ME), region non-uniformity (RN) and relative foreground area error (RFAE), respectively. (C) 2009 Elsevier Ltd. All rights reserved.
Fri, 01 Jan 2010 00:00:00 GMThttp://hdl.handle.net/11455/381272010-01-01T00:00:00ZSTATISTICAL AND SVM-BASED ONCOGENE DETECTION OF HUMAN CDNA EXPRESSIONS FOR OVARIAN CARCINOMAhttp://hdl.handle.net/11455/38131標題: STATISTICAL AND SVM-BASED ONCOGENE DETECTION OF HUMAN CDNA EXPRESSIONS FOR OVARIAN CARCINOMA
作者: Lai, C.H.; 蔡孟勳; Yu, S.S.; Chang, J.D.; Tsai, M.H.; 喻石生
摘要: In this paper, the human ovarian cDNA expression database is analyzed for discriminating oncogenes according to different pathological stages of ovarian carcinoma. The human ovarian cDNA expression database of this paper collects 41 patient samples, which includes 13 samples at normal ovarian tumors, 6 samples of borderline of cancers, 7 samples of ovarian cancer at stage I and 15 samples of ovarian cancer at stage III. Due to 9,600 genes of each pathological sample, a large number of genes are analyzed and discovered difficulty. For this reason, linear regression and analysis of variance (ANOVA) are used to discover and detect 21 notable oncogenes. Further more, these 21 notable oncogenes are divided and examined by support vector machine (SVM) with 5 different classifications according to their gene expressions of pathological stages. From the experimental results, the average accuracy of 5 classification experiments is 89% in cross validation. Moreover, the related scientific literatures also indicate these 21 discovered oncogenes are related to ovarian cancer or other cancers. It proves these discovered oncogenes are highly related to different cancers. Finally, this paper also develops a graphical user interface (GUI) bio-statistical system for gene expression analysis to assist doctors and pathologists to analyze and diagnose ovarian cancer.
Thu, 01 Jan 2009 00:00:00 GMThttp://hdl.handle.net/11455/381312009-01-01T00:00:00Z質式字的研究http://hdl.handle.net/11455/49892標題: 質式字的研究; A Study of Primitive Words
作者: 喻石生
摘要: 設X為一有限的字母集並設X/sup */是由集合X所 生成之自由么半�.令X/sup +/=X/sup */□.lbrlce. 1 .rbrace. 其中1是空字.X/sup */的元素稱為字,而X/sup * /的任一個子集合都稱為一個言語.當一個字X .in.X /sup X/滿足X=U/sub 1/U/sub 2/......Un,U/sub 1/=U/sub2/...... =U/sub n/ .in. X/sup +/必定導致n=1,則我們稱X為一個 質式字,若把每一個字都看成是一個特別的圖樣( Pattern),則一個質式圖樣表示它不會是一個基本圖 樣重複多次而構成的.我們已經知道如果兩個不 相同的質式字p與q使得pq/sup n/是非質式字對於某 些n .gtoreq. 1,則導致對於所有的k .gtoreq. 2,pq/sup n+k /都會是質式字,但是當pq/sup n/是非質式字時,pq/sup n+1/會形成質式字或非質式字,卻一直是個未知的 問題,許多關係到質式字的研究皆在遇到這個問 題時產生無比的困擾.本文在這方面多年的研究 深覺有必要將此問題作一番深入的研究.
Sat, 01 Jan 1994 00:00:00 GMThttp://hdl.handle.net/11455/498921994-01-01T00:00:00ZDNA相關的字對研究http://hdl.handle.net/11455/51255標題: DNA相關的字對研究; A Study of DNA Related Word Pairs
作者: 喻石生; 趙有光; 蕭鴻貴
Thu, 01 Jan 2004 00:00:00 GMThttp://hdl.handle.net/11455/512552004-01-01T00:00:00Z錯誤偵測碼研究http://hdl.handle.net/11455/51294標題: 錯誤偵測碼研究; A Study of Error Detecting Codes
作者: 喻石生; 黃政治
摘要: 頻道可解碼的數碼模式概念是在[2], [3] 兩篇文章中所提出的，這包含資訊在具有替代、刪除或插入所產生錯誤的頻道中，要達成資訊的解碼與更正時之需求。在一個非同步的系統中，可能會有資料遺失。資訊在有雜訊的無線頻道中傳遞，資訊亦會有減少或增加的可能，而不復維持原資訊的長度。這種錯誤可用刪除或插入的模式來呈現。在此頻道模式下，數碼的錯誤偵測需求，和傳統由替代所產生錯誤的模式下之需求大不相同。在偵測由刪除或插入所產生之錯誤時，一個最簡單的編碼方法是將資訊重覆的傳送。但在[1]一文中證明，重覆五次或更多次時，將資訊重覆傳送的編碼模式並不佳。運用同長度的數碼，來偵測由刪除或插入所產生之錯誤的技術，曾在[5], [6]兩篇文章中被研究到。在[2]一文中提出包含替代、刪除或插入所產生錯誤的抽像頻道模式，作為研究資訊傳輸的編碼問題之一般解決方案。一些關於數碼之抽像頻道的頻道可解碼性質，在[2], [3] 兩篇文章中曾被研究到。[4]中對混合三種基本錯誤型態的抽像頻道，曾作更進一步的分析。[5]中對有限長度序列之錯誤偵測，討論許多，但未討論無限長度序列之錯誤偵測。在此抽像頻道模式下，我們將推導一些無限長度序列之錯誤偵測碼之代數性質。
Tue, 01 Jan 2008 00:00:00 GMThttp://hdl.handle.net/11455/512942008-01-01T00:00:00Z數碼性質研究http://hdl.handle.net/11455/51283標題: 數碼性質研究; A Study on the Properties of Codes
作者: 喻石生; 趙有光
摘要: 頻道可解碼的數碼模式概念是在[2], [3] 兩篇文章中所提出的，這包含資訊在具有替代、刪除或插入所產生錯誤的頻道中，要達成資訊的解碼與更正時之需求。在一個非同步的系統中，可能會有資料遺失。資訊在有雜訊的無線頻道中傳遞，資訊亦會有減少或增加的可能，而不復維持原資訊的長度。這種錯誤可用刪除或插入的模式來呈現。在此頻道模式下，數碼的解碼需求與錯誤的更正需求，和傳統由替代所產生錯誤的模式下之需求大不相同。在偵測由刪除或插入所產生之錯誤時，一個最簡單的編碼方法是將資訊重覆的傳送。但在[1]一文中證明，重覆五次或更多次時，將資訊重覆傳送的編碼模式並不佳。運用同長度的數碼，來偵測由刪除或插入所產生之錯誤的技術，曾在[5], [6]兩篇文章中被研究到。在[2]一文中提出包含替代、刪除或插入所產生錯誤的抽像頻道模式，作為研究資訊傳輸的編碼問題之一般解決方案。一些關於數碼之抽像頻道的頻道可解碼性質，在[2], [3] 兩篇文章中曾被研究到。[4]中對混合三種基本錯誤型態的抽像頻道，曾作更進一步的分析。在此抽像頻道模式下，我們將推導一些頻道可解碼與頻道可更正的數碼之代數性質，並提出一種頻道可解碼與可更正的數碼。
Sun, 01 Jan 2006 00:00:00 GMThttp://hdl.handle.net/11455/512832006-01-01T00:00:00Z自動機性質的研究http://hdl.handle.net/11455/51290標題: 自動機性質的研究; A Study on the Properties of Automata
作者: 喻石生; 黃政治
摘要: 考慮XML資料串流中，大量XPath表示式的比對問題。使用確定性自動機(DFA)是其中的一種方法。確定性自動機能夠有效率的解決比對的問題：根據一般的經驗，我們可以穫得穩定的結果，而無關乎XPath表示式的個數。這在大量的查詢中，非常的重要。由於網際網路的快速成長，越來越多的網路資料與應用程式使用XML規格來儲存資料。但是，XML文件是屬於半結構化的資料，因此，一種新的資料存取比對方式，用以處理XML資料是非常必要的。此外，我們也將研究字串比對的演算法，利用確定性自動機，來由文件中搜尋多族關鍵字串，並標示相對位置。這些演算法包含，比對用確定性自動機的建構，及使用所建構之確定性自動機在單一搜尋中，去對文件作多族關鍵字串的比對。因此，這些比對用確定性自動機，不只用以同時於文件中搜尋多族關鍵字串的出現與否，更將標示出多族關鍵字串出現的順序。當然，關於多族比對用確定性自動機的最小化，我們也將加以研究。相關的性質，亦是我們將探討的課題。
Mon, 01 Jan 2007 00:00:00 GMThttp://hdl.handle.net/11455/512902007-01-01T00:00:00Z字首質式字的研究http://hdl.handle.net/11455/51278標題: 字首質式字的研究; A Study of Prefix Promitive Words
作者: 喻石生; 趙有光; 蕭鴻貴
摘要: 在[Wu92]一文中，他們考慮能保持d-質式言語的同態對映之性質。而[Lot97]一書中則討論到關於能保持無平方字的同態對映之特徵表示。類似這些研究，提供了研究能保持某種性質的同態對映的原始動機。Shyr 及Thierrin 在[Shyr77]中提出了能保持質式字的同態對映的一些性質，他們證明了：如果一個嵌射同態對映h: X* → X*使得h( X) 形成一個純碼，則h能保持質式字的質式性。理論上的特徵表示或性質，就像這個保持質式字的質式性之同態對映的性質，有時不能有效的降低檢查一個同態對映是否能保持某種性質時的複雜度。更實際有效的方式是提供一個能在有限步驟中判定一個同態對映是否能保持某種性質的方法。有許多學者致力於研究能保持某種性質之同態對映的這種特徵。例如：Crochemore 在[Cro82]一文中，提出一個方法能在有限步驟中檢驗同態的無平方保持性。Hsiao 等在[Hsi03]一文中，提出另一個更有效的方法。此計畫規劃如下：(i)定義及介紹必要之概念、(ii)找出一個能在有限步驟中檢驗同態對映是否保持字首質式性的方法、(iii)驗證能保持字首質式性的同態對映與其它性質保持同態對映間的關係、(iv)研究字首n-次方字的性質。
Sat, 01 Jan 2005 00:00:00 GMThttp://hdl.handle.net/11455/512782005-01-01T00:00:00ZDNA 相關言語的研究http://hdl.handle.net/11455/49692標題: DNA 相關言語的研究; A Study of DNA Related Languages
作者: 喻石生
Tue, 01 Jan 2002 00:00:00 GMThttp://hdl.handle.net/11455/496922002-01-01T00:00:00Zμ-消去言語的性質http://hdl.handle.net/11455/49684標題: μ-消去言語的性質; Properties of .mu.-Reducible Languages
作者: 喻石生
Sat, 01 Jan 2000 00:00:00 GMThttp://hdl.handle.net/11455/496842000-01-01T00:00:00ZA Note on d-Primitive Languageshttp://hdl.handle.net/11455/49676標題: A Note on d-Primitive Languages; D-質式言語研究
作者: 喻石生
摘要: 在本計劃中，我們打算研究�-消去正規言語、�-消去文脈無關言語、�-消去區域言語及�-消去數碼。我們也打算找出某些言語的�-消去子的特徵。本研究計劃的成果將可提供言語乘積與質式字次方之間關係的一個深入的描述。
Fri, 01 Jan 1999 00:00:00 GMThttp://hdl.handle.net/11455/496761999-01-01T00:00:00ZLibrary Service(II)http://hdl.handle.net/11455/49700標題: Library Service(II); 數學研究推動中心中部地區圖書服務計畫(II)
作者: 喻石生
Thu, 01 Jan 2004 00:00:00 GMThttp://hdl.handle.net/11455/497002004-01-01T00:00:00ZA Study of Annihitating Functionshttp://hdl.handle.net/11455/49660標題: A Study of Annihitating Functions; 消去函數的研究
作者: 喻石生
摘要: 令X是一個有限的字母集且令X/sup */為X所生 成的自由模,任一個X/sup */的元素都稱為"字,令 X/sup +/=X/sup */\{ 1} 其中1指空字,而任一個X/sup * /的子集都稱為言語,一個字x .in. X/sup+/凡存在 有 .mu./sub 1/=...=.mu./sub n/ .in. X/sup +/使得x=.mu. /sub1/....mu./sub n/則導至n=1,此時稱x為一個"質字 ,令Q為所有質字的集合.對2X/sup */上一個給定 的運算□及一給定的言語集合F,F-消去函數,表 示成.alpha.F,是一個定義為 .alpha.F(A)=A的函數,其 中A .in. X/sup */,A則是所有言語B的集合,而B滿足A □B .in. F且如果A□C .in. F且 B C則導至B=C.例如, 設□為言語的乘積運算且令q為Q的所有非空子 集的集合,則.alpha.q是本人和石煇然教授合著之 "言語消去子"一文中所稱之消去函數,在這裡我們稱之為q-消去函數.本研究計畫的目的是在研 究對於任意給定之運算□及任一給定之言語的 集合F時,F-消去函數的一般化之性質,我們也將 研究對於特別的言語集合以及特殊的運算時之 消去函數的性質.例如,在"言語消去子"一文中, 對於任一言語A,.alpha.q(A)只包含一個集合,因此 我們能令.alpha.q(A)=B其中B為某一言語,但是這 一個性質一般情況而言,並不成立,諸如此類的 許多性質正等著我們去研究,及找出真相,這個 研究將可以節省我們去研究特例的時間及精力 .
Sun, 01 Jan 1995 00:00:00 GMThttp://hdl.handle.net/11455/496601995-01-01T00:00:00ZW-言語的研究http://hdl.handle.net/11455/49688標題: W-言語的研究; A Study of W-Languages
作者: 喻石生
Mon, 01 Jan 2001 00:00:00 GMThttp://hdl.handle.net/11455/496882001-01-01T00:00:00Z分佈於{0,1}+空間的錯誤更正碼之研究http://hdl.handle.net/11455/51561標題: 分佈於{0,1}+空間的錯誤更正碼之研究; A Study on Error-Correcting Codes in {0,1}+
作者: 喻石生
摘要: Concepts of channel-decodable codes were proposed in [3] and [4], which concern the requirements of the decodability and error-correctability of messages including errors caused by substitutions, deletions, or insertions. When a system is not synchronized, some segments of messages could be lost. Or, when we transmit messages through a noisy wireless channel, messages could be changed by deletions or insertions and no more keep the lengths of original messages. These errors can be modeled as symbol deletions or insertions. An easiest way in coding applications to detect the deletion or insertion errors is to use a code repeatedly to transmit a sequence of messages. But it is shown [1] that the coding scheme obtained by the repeated use of codes is bad, if used five times or more. Techniques to detect or to correct the bit insertion or deletion errors in messages coded by uniform codes have been studied in [6] and [7].As a general approach to investigate the decoding problems in message transmission, a model of abstract channels with insertion, deletion, or substitution errors was proposed in [3]. Some properties required for a code to be channel-decodable for a given abstract channel were studied in [3] and [4]. A further analysis of these abstract channels that involve combinations of three basic error types was given by Konstantinidis [5]. In [6], the error-detecting for finite length messages is studied. But for the error in infinite length messages (general case), the error detectable codes has not been investigated. Following this model of channels, some algebraic properties of channel-detectable and -correctable codes have been derived in [2]. A kind of channel-detectable and -correctable codes is also proposed in [2]. But more general properties of channel-correctable codes and maximal channel-correctable codes still have to be further investigated.[1] K. A. S. Abdel-Ghaffar, Repeated Use of Codes for Error Detection Five Times is Bad, IEEE Trans. Inform. Theory, Vol.48, No.7 (2002), 2053--2060.[2] H.K. Hsiao, Hsiu-Hsia Lin and S.S. Yu, The Decodability and Correctability of Codes, Int. J. Computer Math., (Accepted).[3] H. Jürgensen and S. Konstantinidis, Variable-Length Codes for Error Correction, Automata, languages and programming (Szeged, 1995), 581--592, Lecture Notes in Comput. Sci., 944, Springer, Berlin, 1995.[4] H. Jürgensen and S. Konstantinidis, Error Correction for Channels with Substitutions, Insertions, and Deletions, Information theory and applications, II (Lac Delage, PQ, 1995), 149--163, Lecture Notes in Comput. Sci., 1133, Springer, Berlin, 1996.[5] S. Konstantinidis, Structural Analysis of Error-Correcting Codes for Discrete Channels That Involve Combinations of Three Basic Error Types, IEEE Trans. Inform. Theory, Vol.45, No.1 (1999), 60--77.[6] S. Konstantinidis and Amber O'Hearn, Error-Detecting Properties of Languages, Theor. Comp. Sci., Vol.276 (2002), 355—375.[7] V. I. Levenshtein, Binary Codes Capable of Correcting Deletions, Insertions, and Reversals, Soviet Phys. Dokl., Vol.10, No.8 (1966), 707--710 (English translation).; 頻道可解碼的數碼模式概念是在[3], [4] 兩篇文章中所提出的，這包含資訊在具有替代、刪除或插入所產生錯誤的頻道中，要達成資訊的解碼與更正時之需求。在一個非同步的系統中，可能會有資料遺失。資訊在有雜訊的無線頻道中傳遞，資訊亦會有減少或增加的可能，而不復維持原資訊的長度。在偵測由刪除或插入所產生之錯誤時，一個最簡單的編碼方法是將資訊重覆的傳送。但在[1]一文中證明，重覆五次或更多次時，將資訊重覆傳送的編碼模式並不佳。運用同長度的數碼，來偵測由刪除或插入所產生之錯誤的技術，曾在[6], [7]兩篇文章中被研究到。在[3]一文中提出包含替代、刪除或插入所產生錯誤的抽象頻道模式，作為研究資訊傳輸的編碼問題之一般解決方案。一些關於數碼之抽象頻道的頻道可解碼性質，在[3], [4] 兩篇文章中曾被研究到。[5]中對混合三種基本錯誤型態的抽象頻道，曾作更進一步的分析。[6]中對有限長度序列之錯誤偵測，討論許多，但未討論無限長度序列之錯誤偵測。在此抽象頻道模式下，一些頻道可解碼與頻道可更正的數碼之代數性質，曾在[2]中被推導出來。在[2]中並提出一種頻道可解碼與可更正的數碼。然而更廣泛的頻道可更正數碼的性質，與更一般化的極大頻道可更正數碼，仍有待更進一步的研究與探討。
Fri, 01 Jan 2010 00:00:00 GMThttp://hdl.handle.net/11455/515612010-01-01T00:00:00ZA Study on Bone Age Assessment System Based on Phalangeal Features of Hand Rediographshttp://hdl.handle.net/11455/51607標題: A Study on Bone Age Assessment System Based on Phalangeal Features of Hand Rediographs; 以指骨特徵為主之手掌X光影像骨齡估測系統之研究
作者: 喻石生
摘要: There are different ROI (region of interest) processing and feature extraction in a computer-assisted bone age assessment (CABAA) system for various stages of skeletal development [1] and these methods are critical points for image analysis to get a fine BAA. The first object is studying and analyzing the ROI processing scheme in a CABAA system. In the previous studies, we proposed a bone age and feature cluster estimation method which applies only a few presentable features extracted from input hand radiographs [2]. After the automatic ROI feature clustering and extracting, one still needs to construct a precise and complete bone age assessment system for features and data in different bone development stages. The aim is to develop an automatic bone age assessment system to provide a higher performance and precision of bone age assessment.Following our previous studies and results, we propose a three year project to progress the study of bone age assessment based on phalangeal features of hand radiographs. The first year study will be focused on methods to construct a bone age assessment system of the early bone development stage. The principal characters of the early bone development stage are the geometric shapes, areas and relative distances, and so on, of epiphyses and metaphyses. These features can be chosen and extracted according to the feature clusters obtained by our previous studies of ROI image segmentation, feature extraction, and bone feature clustering of early bone development stage [3, 4, 5]. Then for hand images in the early stage, we are going to design a suitable fuzzy neural network (FNN) algorithm [6, 7, 8, 9] to assess bone ages according to these features. These results obtained by applying this FNN to our database will be compared with bone ages estimated by clinical doctors. Then this FNN algorithm will be modified and trained again and again [9, 10, 11, 12].The second year project is focused on the feature analysis of ROI images in later stage of bone development. In the later stage, the principal characters of phalanges are the fusion degrees of epiphseal/metaphyseal regions. We will apply wavelet transform and singular value division (SVD) techniques to analyze these characters of fusion degrees. First, a region containing epiphysis and metaphysis is extracted from an ROI image to be a new ROI image. Then one-level wavelet transform is applied to this ROI image to get 4 one-level wavelet frequencies. The energy of each normalized frequency region is calculated and finally an estimation parameter λ is produced as a character value to analyze the texture and get the corresponding character based features in description of epiphyseal fusion. The object of the third year study is the construction and analysis of the bone age assessment of the later stage. According to the results obtained in the second year study, one is going to analyze and extract the features of ROI images in the later stage. Features in the related feature cluster suggested from the results by previous bone-cluster estimation are applied to train a support vector machine (SVM). After training, one can get an SVM for bone age classification and assessment of ROI images in the later stage. Again, the assessment results of ROI images in our database will be compared with bone ages estimated by clinical doctors. The SVM algorithm will be modified and trained again and again. The final goal of our project is to construct a more reliable and flexible fully automatic bone age assessment system.; 在電腦化骨齡估測(CABAA: Computer-assisted Bone Age Assessment) 系統中，有關影像感興趣區域(ROI: Region Interest)處理及特徵選取方法，會因每個階段骨骼發展特性的不同而存在極大的差異性[1]，且往往為最終預測骨齡的重要關鍵。因此，上述相關演算法的分析與研究自然是CABAA系統中的首要任務。在之前的研究中，我們提出了一個可利用少數具代表性的特徵對於待測之手掌影像進行前置的骨齡群集估測的方法[2]。在ROI特徵分類及萃取後，最後所需要的是針對不同時期的特徵及資料，建立正確、完善的骨齡估測系統。目的在於致力發展一套自動化骨齡估測系統，以提升骨齡評估之工作效率及準確性。我們將延續先前的研究計畫及成果，提出三年計劃，來進行以指骨特徵為主之手掌X光影像骨齡估測系統之研究。第一年計畫重點為建立骨骼發展前期(early stage)之骨齡估測方法，因前期指骨骨骼發展主要特徵為骨骺及幹骺端之幾何形狀、面積及相對距離…等項目，此部分可依據我們先前對於前期骨骼發展之ROI影像切割、特徵萃取、骨齡群集估測以及各群集所建議的特徵項目等研究所得到結果[3, 4, 5]，對於骨骼發展前期之影像，設計適當之模糊類神經網路(FNN: Fuzzy Neural Network)演算法[6, 7, 8, 9]進行骨齡估測，最後與臨床醫師之判讀進行交叉分析比較，反覆進行演算法的修正與精進[9, 10, 11, 12]。第二年計畫重點為後期骨骼發展階段的特徵分析，分析後期指骨骨骼發展主要特徵為骨骺與幹骺端融合區塊的融合程度，我們將使用小波轉換(wavelet transform)及奇異值分解(SVD)等技術進行骨骼融合特徵分析。首先從ROI影像中擷取出包含骨骺及幹骺區域作為新的ROI影像，然後以一階小波轉換於新的ROI影像取得四份一階小波頻帶，接著計算各頻帶之正規化區域能量並最後產生估測參數λ值作為特徵值進行紋理分析。第三年計畫為骨骼發展後期(later stage)之骨齡估測系統之建立與分析，依據我們第二年對於後期骨骼發展之ROI影像進行特徵分析與萃取、以及先前得到骨齡群集估測結果與各群集所建議的特徵項目，利用支援向量機進行學習與分類，進而得到後期骨齡估測結果。同樣地，最後再與臨床醫師之判讀進行交叉分析比較，以進行演算法的修正與精進。最終目標為提供一個更具信賴度且彈性的自動化骨齡估測系統。
Sun, 01 Jan 2012 00:00:00 GMThttp://hdl.handle.net/11455/516072012-01-01T00:00:00Z交換質式言語的性質http://hdl.handle.net/11455/51537標題: 交換質式言語的性質; Properties of Abelian Primitive Languages
作者: 喻石生
摘要: Properties of Abelian Primitive LanguagesLet the alphabet X be indexed as X={a0,a1,…,an}. For x∈X* and aj∈X, 0≤j≤n, let #j(x) denote the number of occurrences of aj. The Parikh vector of x, denoted by ψ(x)=(#0(x), #1(x), #2(x), …, #n(x)). An abelian k-power is a non-empty word of the form x1x2…xk, where ψ(x1)= ψ(x2)= …=ψ(xk). If a word is abelian 2-power, then it is called an abelian square. The avoidability concerned abelian squares are investigated in [1]. A non-empty word is called abelian primitive if it is not an abelian k-power for any k≥2. Let Ap be the set of all abelian primitive words over an alphabet X. This project is devoted to investigate the properties of Ap.A language L1⊆X* is said to pseudo-commute with L2 under f : X*→X* iff L1L2= f(L2)L1 [4]. Of special interest from the DNA computing perspective is the special case where f is an antimorphic involution modeling the Watson-Crick complementarity of DNA strands (see [2]). Two languages L1 and L2 over an alphabet X commute if L1L2=L2L1 [5,6]. If two languages L1 and L2 commute and L2 is singleton, then the word in L2 and all words in L1 have the same primitive root [5]. While the properties of abelian primitive words are greatly related to properties of cyclic permutation primitive words and the solutions of the equation L1L=LL1.[1] Keranen, Abelian Squares are Avoidable on 4 Letters, Springer LNCS 623 (1992), 41--52.[2] M. Daley and L. Kari, DNA Computing: Models and Implementations, Comments on Theoretical Biology, 7(3) (2002), 177—198.[3] R.C. Lyndon and M.P. Schutzenberger, The Equation aM=bNcP in a Free Group, Michigan Math. J., 9 (1962), 289--298.[4] L. Kari, K. mahalingam, and S. Seki, On the Pseudo-Commutativity of Words and Languages, (submitted), 2007.[5] H.J. Shyr, On Two Languages That Commute, Note on Semigroup 9, K. Marx. University, Economics, Budapest, (1983), 257—269.[6] S.S. Yu, Languages and Codes, Lecture Notes, Tsang Hai Book Publishing Co., Taichung, Taiwan 2005.; 交換質式言語的性質令字母集X為有序集合X={a0,a1,…,an}。對於x∈X*及aj∈X，0≤j≤n，令#j(x)表示字母aj在x中出現的次數。x的Parikh向量則定義為ψ(x)=(#0(x), #1(x), #2(x), …, #n(x))。一個交換k-次元是一個非空的字具有x1x2…xk的型式，其中ψ(x1)= ψ(x2)= …=ψ(xk)。如果一個字是交換2-次元，則稱之為一個交換平方字。在[1]中是研究關於交換平方字的可避免性。當一個非空的字不是任何k≥2的交換k-次元，則稱之為一個交換質式字。令Ap表示建構於字母集X的所有交換質式字的集合。我們將在此計畫中研究關於Ap的性質。給一個對映f : X*→X*，一個言語L1⊆X*稱為與另一個言語L2為pseudo- commute就是說它們滿足等式L1L2= f(L2)L1 [4]。其中，特別由DNA計算透視所導致得吸引人之處，是在於f為一個非同構次方用以塑造DNA双旋股的Watson-Crick互補模式的特殊狀況(參考[2])。當建構於字母集X的兩個言語L1與L2滿足L1L2=L2L1時，稱它們為可交換。當L1與L2為可交換且L2為含單一元素時，L2中的那個字及所有L1中的字皆有同一個質式根[5]。而交換質式字的性質與旋轉交換質式字及言語等式L1L=LL1的解之性質息息相關。[1] Keranen, Abelian Squares are Avoidable on 4 Letters, Springer LNCS 623 (1992), 41--52.[2] M. Daley and L. Kari, DNA Computing: Models and Implementations, Comments on Theoretical Biology, 7(3) (2002), 177—198.[3] R.C. Lyndon and M.P. Schutzenberger, The Equation aM=bNcP in a Free Group, Michigan Math. J., 9 (1962), 289--298.[4] L. Kari, K. mahalingam, and S. Seki, On the Pseudo-Commutativity of Words and Languages, (submitted), 2007.[5] H.J. Shyr, On Two Languages That Commute, Note on Semigroup 9, K. Marx. University, Economics, Budapest, (1983), 257—269.[6] S.S. Yu, Languages and Codes, Lecture Notes, Tsang Hai Book Publishing Co., Taichung, Taiwan 2005.
Thu, 01 Jan 2009 00:00:00 GMThttp://hdl.handle.net/11455/515372009-01-01T00:00:00ZA Study of Bone Age Clustering System of Hand Radiographshttp://hdl.handle.net/11455/51592標題: A Study of Bone Age Clustering System of Hand Radiographs; 手掌X光影像之骨齡分群系統之研究
作者: 喻石生; 徐山靜
摘要: 手掌X 光影像之骨齡分群系統之研究在電腦化骨齡估測(CABAA: Computer-assisted Bone Age Assessment) 系統中，有關影像感興趣區域(ROI: Region Interest)處理及特徵選取方法，會因每個階段骨骼發展特性的不同而存在極大的差異性[1]，且往往為最終預測骨齡的重要關鍵。因此，上述相關演算法的分析與研究自然是CABAA 系統中的首要任務。然而，待測影像究竟屬於哪一個骨骼發展階段？在先前有關於自動化骨齡估測的文獻中往往被視為所謂的先驗知識(a priori knowledge)[2, 3, 4, 5]，用來作為選擇ROI 分析技術及分類器的依據。為了避免這種必須依賴經驗去預先進行判斷骨骼階段的問題，可在選擇估測骨齡的分類器之前，利用少數具代表性的特徵對於待測之手掌影像進行前置粗略的骨齡分群。藉由分析上述分群結果進而提出一個新的骨骼發展階段界定，以及各階段相對的ROI 分析技術(如幾何特徵切割或骨骼融合分析)與分類器。如此，不僅可提升骨齡評估之效率外，還可提供一個更具信賴度且彈性的自動化骨齡估測系統。在此我們提出三年計劃，來進行手掌X 光影像之骨齡分群之研究。第一年計畫重點為ROI 影像幾何特徵切割，先將指骨ROI 影像進行正規化之後，再依據我們在先前的研究論文中提出的影像切割方法[6, 7]對於ROI 影像進行切割處理，最後與臨床醫師之判讀進行交叉分析比較[8]，計算相對於理想狀況之正確切割比率，反覆進行演算法之修訂及創新，以期達到精確之影像幾何特徵切割結果，作為早期骨骼發展階段的主要特徵萃取之依據。第二年計畫為骨骼融合程度分析，依據TW2 [9]骨齡估測法，後期骨骼發展階段的特徵為骨骺與幹骺端融合區塊的融合程度，我們將透過影像頻率域訊號處理的方法如小波轉換(Wavelet transform)及灰階共生矩陣(Co-occurrence Matrices)等技術[10,11]，進行骨骼融合特徵分析。第三年計畫為骨齡分群系統之建立與分析，依據第一年計畫產生之ROI影像幾何切割處理結果，分析骨骺部份在各個骨齡階段中之發展特性，找出具鑑別性的骨骺幾何特徵，再運用模糊類神經網路(FNN: Fuzzy Neural Network)演算法[12 13,14, 15]進行骨齡分群。最後，分析由FNN 系統所得到之骨齡分群結果，定義出不同於傳統的骨骼發展階段之界定，以及建議各階段所對應的ROI 處理及特徵選取方法(依據前兩年提出的結果)以作為最終骨齡估測分類器選取之依據。; A study of bone age clustering system of hand radiographs.There are different ROI (region of interest) processing and feature extraction in a computer-assistedbone age assessment (CABAA) system for various stages of skeletal development [1] and these methodsare critical points for image analysis to get a fine BAA. The first object is studying and analyzing the ROIprocessing scheme in a CABAA system. Nevertheless, the related modeling techniques are variousdepending on the characteristics of different stages of bone maturity which usually is taken as an a prioriknowledge in most previously proposed schemes [2, 3, 4, 5]. In order to avoid pre-determining the testimage belongs to which stage by experience. If a coarse bone age cluster for a hand radiograph could beassigned, by using some presented features then these corresponding image analysis methods (such asgeometric segmentation and fusion degree) employed for feature analysis can be identified automatically.This could give a promotion of efficiency for BA estimation and provide a more flexible BAA system.In this project, we will propose three-year study to investigate the bone age clustering analysis for handradiograph. The first year study will focus ROI preprocessing, which includes will ROI normalization andROI segmentation. The segmentation region of epiphyseal/metaphyseal in an EMROI could be obtainedusing the modified adaptive 2-means clustering method proposed in [6, 7] for separating epiphysis andmetaphysis from the soft tissue. Then compare with the clinician's interpretation [8] repeatedly andcalculate the correct segmentation rate to the ideal state to revise the algorithms of performing ofabove-mentioned algorithm, in the hope of reaching the result of accurate image geometry featuresegmentation. The second year project will examine the fusion degree analysis. According the TW2 method[9], once the size does not differ, the degree of fusion must be extracted. A wavelet analysis is used whenthe epiphyseal fusion has started in the later stage. We will apply the wavelet transform and Co-occurrenceMatrices [10, 11] to get the corresponding characteristics based features in description of epiphyseal fusion.The object is the construction and analysis using the fuzzy neural network (FNN) [12, 13, 14, 15] for boneage clustering system in last year study. From the observation of the previously segmented results, somecharacteristics of epiphyseal/metaphyseal defined as the features which are sufficient for the featureextraction. According to the extracted features analysis and the knowledge of expert [9], the rules of fuzzyinference are constructed. Then, an artificial neural network is used to form a learning system. Then theoutput of the presented FNN system is a bone age cluster, which is a coarse range of bone age. Finally,bone clustering analysis for the method of corresponding ROI processing and feature extraction isdiscussed based on the prediction results of FNN system. By this analysis, a new defined stages for skeletaldevelopment and the corresponding image analysis method are given to provide a more flexible andreliable BAA system.
Sat, 01 Jan 2011 00:00:00 GMThttp://hdl.handle.net/11455/515922011-01-01T00:00:00Z數學研究推動中心中部地區圖書服務計畫(I)http://hdl.handle.net/11455/49914標題: 數學研究推動中心中部地區圖書服務計畫(I); Library Service (I)
作者: 喻石生
Wed, 01 Jan 2003 00:00:00 GMThttp://hdl.handle.net/11455/499142003-01-01T00:00:00ZA HDWT-based reversible data hiding methodhttp://hdl.handle.net/11455/38150標題: A HDWT-based reversible data hiding method
作者: Chan, Y.K.; 蔡垂雄; Chen, W.T.; Yu, S.S.; Ho, Y.A.; Tsai, C.S.; Chu, Y.P.; 喻石生; 詹永寬
摘要: This paper presents a reversible data hiding method which provides a high payload and a high stego-image quality. The proposed method transforms a spatial domain cover image into a frequency domain image using the Haar digital wavelet transform (HDWT) method, compresses the coefficients of the high frequency hand by the Huffman (or arithmetic) coding method, and then embeds the compression data and the secret data in the high frequency hand. Since the high frequency hand incorporates less energy than other bands of an image, it can be exploited to carry secret data. Furthermore, the proposed method utilizes the Huffman (or arithmetic) coding method to recover the cover image Without any distortion. The proposed method is simple and the experimental results show that the designed method can give a high hiding capacity with a high quality of stego-image. (C) 2008 Published by Elsevier Inc.
Thu, 01 Jan 2009 00:00:00 GMThttp://hdl.handle.net/11455/381502009-01-01T00:00:00Z後部+及後段*言語http://hdl.handle.net/11455/49668標題: 後部+及後段*言語; Post-Plus and Suffix-Star Languages
作者: 喻石生; 石煇然
Wed, 01 Jan 1997 00:00:00 GMThttp://hdl.handle.net/11455/496681997-01-01T00:00:00Z數碼之代數分類http://hdl.handle.net/11455/49663標題: 數碼之代數分類; Some Algebraic Classifications of Codes
作者: 石煇然; 喻石生
摘要: 請參見八十三年十二月出版之"進行中研究 計畫-理工類"P.714該篇摘要.
Mon, 01 Jan 1996 00:00:00 GMThttp://hdl.handle.net/11455/496631996-01-01T00:00:00Z自行插入質式字的研究http://hdl.handle.net/11455/49697標題: 自行插入質式字的研究; A Study of Self-Inserted Primitive Words
作者: 喻石生; 趙有光; 蕭鴻貴
Wed, 01 Jan 2003 00:00:00 GMThttp://hdl.handle.net/11455/496972003-01-01T00:00:00ZA PCA AND PERTURB BASED FRAGILE WATERMARKING SCHEME FOR 3D MODELShttp://hdl.handle.net/11455/38000標題: A PCA AND PERTURB BASED FRAGILE WATERMARKING SCHEME FOR 3D MODELS
作者: Wang, J.T.; 喻石生; Wang, P.C.; Yu, S.S.
摘要: In this paper, a new fragile watermarking scheme that enables the verification of 3D models for detecting unauthorized alterations is presented. The principal component analysis (PCA) is adopted to produce the PCA coordinate system and make the system robust to similarity transformation attacks. Before watermark embedding, the coordinates of illegitimate intervals are slightly perturbed to ensure all intervals legitimate. In order to control distortion, a special quantization technique is proposed to watermark a legitimate interval. Experimental results show that 100% vertices of the cover model can be embedded and extracted in the proposed method. Moreover, the distortion of the stego model is insignificant and under controlled by quantization step setting. Finally, the proposed method is immune to the causality, convergence and embedding holes problems.
Fri, 01 Jan 2010 00:00:00 GMThttp://hdl.handle.net/11455/380002010-01-01T00:00:00ZWord-paired catenations of regular languageshttp://hdl.handle.net/11455/37974標題: Word-paired catenations of regular languages
作者: Hsiao, H.K.; 喻石生; Yu, S.S.; Zhao, Y.K.
摘要: In this paper, we mainly focus on the properties of word-paired catenations of regular languages and related languages. We propose a characterization of local regular languages from the viewpoint of word-paired catenation. We also investigate the properties of languages such that the word-paired catenations are regular-free. The languages, on condition that the word-paired catenations of these languages and their complements or reverses are regular, are studied too. (C) 2004 Elsevier B.V. All rights reserved.
Thu, 01 Jan 2004 00:00:00 GMThttp://hdl.handle.net/11455/379742004-01-01T00:00:00ZPrefix-primitivity-preserving homomorphismshttp://hdl.handle.net/11455/37970標題: Prefix-primitivity-preserving homomorphisms
作者: Huang, C.C.; 喻石生; Yu, S.S.
摘要: This paper aims to investigate homomorphisms which preserve p-primitive languages. A characterization of p-primitivity-preserving homomorphisms can be detected within finite steps. Also the set of square-freeness-preserving homomorphisms is shown to be a proper subfamily of the set of p-primitivity-preserving homomorphisms. For homomorphisms over an alphabet X with vertical bar X vertical bar=2, it is also shown that the set of p-primitivity-preserving homomorphisms is a proper subfamily of the set of primitivity-preserving homomorphisms. But it is conjectured to also hold for homomorphisms over an alphabet with more than two letters. (C) 2007 Elsevier B.V. All rights reserved.
Tue, 01 Jan 2008 00:00:00 GMThttp://hdl.handle.net/11455/379702008-01-01T00:00:00ZA note on the free submonoids of the monoid of prefix codeshttp://hdl.handle.net/11455/37980標題: A note on the free submonoids of the monoid of prefix codes
作者: Shyr, H.J.; 喻石生; Yu, S.S.
摘要: Let A* be a free monoid generated by a finite alphabet A and let A(+) = A*\ {1}, where 1 is the empty word. Let M = {L\L subset of or equal to A(+) or L = {1}} be the monoid of languages under the catenation. We consider {1} as a prefix code. The family of all prefix codes p is a free submonoid of M. For any L is an element of M, let L-L = {L' is an element of M\LL' is an element of P}. We show that L-L boolean OR {1} is a submonoid of p for every non-empty language L subset of or equal to A(+). The purpose of this note is to give a characterization for a finite language L in which L-L boolean OR {1}is a free submonoid of P for the case \A\ greater than or equal to 2. For an infinite language L whether the monoid L-L boolean OR {1} is free or not depends on L. We illustrate the above two cases with examples.
Thu, 01 Jan 1998 00:00:00 GMThttp://hdl.handle.net/11455/379801998-01-01T00:00:00ZOn n-power prefix languageshttp://hdl.handle.net/11455/37984標題: On n-power prefix languages
作者: Fan, C.M.; 喻石生; Huang, C.C.; Yu, S.S.
摘要: A word u is said to be an n-power prefix of another word w if w is an element of u(n) X*. Let P-i denote the family of words which have i-power prefixes but no (i +1)-power prefix. Words in P-1 are called p-primitive words. In this paper, some basic properties of words in sets P-1(i) and Pi are studied. It is shown that the sets P-i, P-i \ Q, P-i boolean AND Q, Q \ P-i, P-1(i) and P-i boolean OR Q for i >= 1 are disjunctive. That is, they are dense non-regular languages. A characterization of words in P-1((I)) for i >= 2 is derived. Several properties of words in P-n \ P-1(n) are considered too.
Tue, 01 Jan 2008 00:00:00 GMThttp://hdl.handle.net/11455/379842008-01-01T00:00:00ZPalindrome words and reverse closed languageshttp://hdl.handle.net/11455/37985標題: Palindrome words and reverse closed languages
作者: Yu, S.S.; 喻石生
摘要: A word which is equal to its mirror image is called a palindrome word. Any language consisting of palindrome words is called a palindrome language. In this paper we investigate properties of palindrome words and languages. We show that there is no dense regular language consisting of palindrome words. A language contains all the mirror images of its elements is called a reverse closed language. Clearly, every palindrome language is reverse closed. We show that whether a given regular or context-free language is reverse closed is decidable. We study certain properties concerning reverse closed finite maximal prefix codes in this paper. Properties of languages that commute with reverse closed languages are investigated too.
Sat, 01 Jan 2000 00:00:00 GMThttp://hdl.handle.net/11455/379852000-01-01T00:00:00ZA note of ins-primitive wordshttp://hdl.handle.net/11455/37977標題: A note of ins-primitive words
作者: Hsiao, H.K.; 喻石生; Yeh, Y.T.; Yu, S.S.
摘要: The aim of this article is to investigate the properties of primitive words with respect to insertion operation, called insprimitive words. This includes the investigation of properties concerning primitive words (with respect to catenation operation) in order to show our result concerning the self-insertion of a word. We also show that the set of all ins-primitive words is not only dense but also disjunctive (hence not regular).
Thu, 01 Jan 2004 00:00:00 GMThttp://hdl.handle.net/11455/379772004-01-01T00:00:00ZFactors associated with operative hysteroscopy outcome in patients with uterine adhesions or submucosal myomashttp://hdl.handle.net/11455/37993標題: Factors associated with operative hysteroscopy outcome in patients with uterine adhesions or submucosal myomas
作者: Chang, C.Y.Y.; 喻石生; Chang, Y.T.; Chien, S.C.; Yu, S.S.; Hung, Y.C.; Lin, W.C.
摘要: Objective: To determine the factors associated with hysteroscopic surgery long-term outcome in patients with intrauterine adhesions or submucosal myomas. Methods: Factors thought to be associated with outcome were retrospectively evaluated from the records of 591 patients who were followed up for at least 5 years after undergoing hysteroscopic adhesiolysis (n=203) or myomectomy (n=388). Results: The major factors affecting outcome were degree of adhesion (OR, 1.91; P=0.03) in the former group and parity (OR, 0.55; P=0.005) and depth of intramural penetration of the myoma (OR, 30.74; P<0.001) in the latter. Severe intrauterine adhesion, low parity, and deep intramural penetration of submucosal myoma had an associated increase risk of poor outcome. The overall complication rate was 1.35% and, respectively, 12.8% and 9.3% of the patients who underwent hysteroscopic adhesiolysis or myomectomy needed a second intervention. Conclusion: Hysteroscopic surgery is a safe and effective procedure. Degree of adhesion or parity and depth of intramural penetration of myomas are the major factors affecting outcome in patients with these lesions. (C) 2009 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All tights reserved.
Fri, 01 Jan 2010 00:00:00 GMThttp://hdl.handle.net/11455/379932010-01-01T00:00:00ZA HIGH VERIFICATION CAPACITY REVERSIBLE FRAGILE WATERMARKING SCHEME FOR 3D MODELShttp://hdl.handle.net/11455/37997標題: A HIGH VERIFICATION CAPACITY REVERSIBLE FRAGILE WATERMARKING SCHEME FOR 3D MODELS
作者: Wang, J.T.; 喻石生; Yang, Y.W.; Chang, Y.T.; Yu, S.S.
摘要: The protection and authentication problems for three dimensional (3D) models have become more and more important. The capability of accurately verifying, locating, and recovering any tampered vertex is one of the most important issues. Published fragile watermarking 3D model schemes have a main drawback, the synchronization problem. To increase the capacity of verification and overcome this drawback, a high verification capacity reversible fragile watermarking scheme for 3D models in the spatial domain is proposed in this paper. Initially, a reversible quantization index modulation embedding scheme employs the principal component analysis (PCA) and modulates the distance from the vertex coordinate to the gravity center of the cover model. And, the spread spectrum technique is adopted to improve the robustness while obeying the sensitive fragility and low distortion requirements for watermarking 3D models. The proposed method can achieve that 100% vertices of the cover model can be watermarked, extracted and verified, while the other 3D fragile watermarking schemes can not. In the meanwhile, the synchronization problem is overcome. Moreover, the proposed method is immune to the causality and convergence problems. Finally, the proposed scheme is robust against vertex reordering and similarity transformation attacks.
Sat, 01 Jan 2011 00:00:00 GMThttp://hdl.handle.net/11455/379972011-01-01T00:00:00ZAN ITERATIVE BASED NOVEL MULTI-NUCLEUS DETECTION SCHEME FOR PROTOZOAN PARASITE MICROSCOPIC IMAGEShttp://hdl.handle.net/11455/38003標題: AN ITERATIVE BASED NOVEL MULTI-NUCLEUS DETECTION SCHEME FOR PROTOZOAN PARASITE MICROSCOPIC IMAGES
作者: Lai, C.H.; 喻石生; Wang, H.Y.; Tsai, Y.C.; Yu, S.S.
摘要: Protozoan parasites cause many diseases, such as malaria, EHEC infection, shigellosis, amoebiasis, etc. Different kinds and growing stages of protozoan parasites would lead to different treatments. The most significant characteristic of different growing stages is the number of nuclei. But some nuclei in a cell could be unclear causing the missing in nucleus detection. Common and traditional segmentation methods can not be used to obtain satisfied results directly. This paper presents a novel multi-nucleus detection scheme which is composed from adaptive protozoan parasite boundary erasure, iterative gamma equalization, two-means clustering algorithm, modified connected component detection method, and circle mask scoring method. Except the two-means clustering algorithm, all other parts are modified methods or new methods designed for nucleus extraction, Experiments show that the proposed scheme can detect the nuclei with indistinct boundaries effectively and can obtain better results than other commonly used image segmentation methods.
Thu, 01 Jan 2009 00:00:00 GMThttp://hdl.handle.net/11455/380032009-01-01T00:00:00Z