本文選題：海圖生產(chǎn)體系框架 + 海圖源數(shù)據(jù)融合模型��； 參考：《解放軍信息工程大學(xué)》2013年博士論文

【摘要】：數(shù)字化解決了數(shù)據(jù)從無到有的問題，，而信息化則需要進(jìn)一步實(shí)現(xiàn)數(shù)據(jù)的融合共享，以達(dá)到數(shù)據(jù)的一致性、實(shí)時(shí)性、高效性。我國(guó)的海圖生產(chǎn)雖然已經(jīng)實(shí)現(xiàn)了數(shù)字化，但是相比于發(fā)達(dá)國(guó)家的海圖生產(chǎn)水平而言，存在著產(chǎn)品數(shù)據(jù)不一致、更新周期相對(duì)較慢和生產(chǎn)效率難以較大提高等問題。針對(duì)這些問題，本文提出了一個(gè)新的海圖生產(chǎn)體系，并對(duì)其關(guān)鍵技術(shù)進(jìn)行了研究。論文主要工作和取得的成果有： 1．對(duì)現(xiàn)代海圖生產(chǎn)過程中涉及到的“海圖生產(chǎn)體系”、“海圖數(shù)據(jù)模型”、“自動(dòng)綜合算法”和“空間數(shù)據(jù)匹配”等四個(gè)關(guān)鍵問題的國(guó)內(nèi)外研究現(xiàn)狀進(jìn)行了評(píng)述和分析。 2．提出了同源多尺度海圖生產(chǎn)體系的概念、框架結(jié)構(gòu)，分析了其主要生產(chǎn)工序和關(guān)鍵技術(shù)。該生產(chǎn)體系通過多源海圖數(shù)據(jù)的融合，保證源數(shù)據(jù)的唯一性、現(xiàn)勢(shì)性和產(chǎn)品數(shù)據(jù)的同宗同源；同時(shí)允許產(chǎn)品數(shù)據(jù)之間存在派生關(guān)系，避免了因源數(shù)據(jù)和產(chǎn)品數(shù)據(jù)的比例尺跨度過大導(dǎo)致數(shù)據(jù)生產(chǎn)的困難；引入了“空間幾何實(shí)例共享池”，實(shí)現(xiàn)不同產(chǎn)品數(shù)據(jù)之間空間幾何數(shù)據(jù)的共享，保證了產(chǎn)品數(shù)據(jù)之間的一致性。 3．針對(duì)實(shí)體模型存在數(shù)據(jù)冗余、容易導(dǎo)致數(shù)據(jù)不一致，而拓?fù)淠Ｐ蛣t過于復(fù)雜以及存在關(guān)系冗余的問題，本文提出了一個(gè)無縫無級(jí)空間數(shù)據(jù)融合模型，不僅能夠?qū)崿F(xiàn)多源空間數(shù)據(jù)的融合，而且避免了傳統(tǒng)空間數(shù)據(jù)模型中的數(shù)據(jù)冗余和不一致等問題；結(jié)合S-100和S-57兩個(gè)海圖數(shù)據(jù)國(guó)際標(biāo)準(zhǔn)，給出了海圖要素模型和海圖屬性模型，提出了“最小拓?fù)洹笨臻g模式；同時(shí)，依據(jù)海圖源數(shù)據(jù)融合模型，提出了一套海圖源數(shù)據(jù)庫設(shè)計(jì)與實(shí)現(xiàn)的方案：通過海圖概念字典和要素目錄實(shí)現(xiàn)海圖數(shù)據(jù)語義的統(tǒng)一，通過要素集合和空間集合的解耦合實(shí)現(xiàn)了海圖源數(shù)據(jù)庫的無縫無級(jí)表達(dá)。 4．針對(duì)外部數(shù)據(jù)與海圖源數(shù)據(jù)不一致的問題，本文提出了一種基于XML，由代詞、名詞、形容詞、條件語句、操作語句等語法結(jié)構(gòu)構(gòu)成的轉(zhuǎn)換語言，實(shí)現(xiàn)了要素和屬性的自動(dòng)轉(zhuǎn)換，并通過拓?fù)渲貥?gòu)技術(shù)實(shí)現(xiàn)了空間模式的自動(dòng)轉(zhuǎn)換，進(jìn)而實(shí)現(xiàn)了DNC和CDC數(shù)據(jù)等外部數(shù)據(jù)向海圖源數(shù)據(jù)的轉(zhuǎn)換；研究了新數(shù)據(jù)與源數(shù)據(jù)庫中要素的自動(dòng)匹配方法，給出了一種“基于誤差理論的改進(jìn)型Hausdorff距離”；最后分析總結(jié)了不同海圖源數(shù)據(jù)之間的質(zhì)量評(píng)價(jià)因子和質(zhì)量評(píng)價(jià)規(guī)則知識(shí)。 5．針對(duì)從源數(shù)據(jù)庫中派生產(chǎn)品數(shù)據(jù)的情況，本文重點(diǎn)研究了海圖水深的自動(dòng)綜合方法。通過對(duì)現(xiàn)有水深點(diǎn)群綜合方法的分析，發(fā)現(xiàn)傳統(tǒng)方法將水深點(diǎn)群視為平面點(diǎn)集、過分強(qiáng)調(diào)空間分布特征，導(dǎo)致水深自動(dòng)綜合難以實(shí)用化；針對(duì)現(xiàn)有水深點(diǎn)群自動(dòng)綜合方法的不足，提出了“將水深點(diǎn)群視為空面點(diǎn)集、以海底地形特征識(shí)別為基礎(chǔ)進(jìn)行背景水深自動(dòng)綜合”的新思路；通過建立Delaunay三角網(wǎng)和局部動(dòng)態(tài)更新技術(shù)，以“離散曲率最小”、“平均間距約束”、“深度變化約束”及其組合方式為約束條件，實(shí)現(xiàn)了基于TIN空間曲面的水深點(diǎn)群漸進(jìn)式自動(dòng)綜合。 6．針對(duì)海圖源數(shù)據(jù)庫出現(xiàn)更新時(shí)，需要將變化信息傳遞至產(chǎn)品數(shù)據(jù)的情況，本文重點(diǎn)分析了產(chǎn)品數(shù)據(jù)中曲線的更新方法。利用曲線約束Delaunay三角網(wǎng)建立CDT-Btree，并通過偽彎曲剔除、重采樣等優(yōu)化以及并列彎曲識(shí)別和調(diào)整等方法，構(gòu)造了一種與視覺感受相一致的曲線彎曲特征CDT-Mtree；以CDT-Mtree為基礎(chǔ)，通過多級(jí)彎曲匹配，實(shí)現(xiàn)了源曲線和產(chǎn)品曲線的彎曲匹配和變化識(shí)別，進(jìn)而實(shí)現(xiàn)了產(chǎn)品曲線中變化彎曲的增量更新。 7．給出了一個(gè)同源多尺度海圖生產(chǎn)的實(shí)驗(yàn)。以DNC和CDC實(shí)驗(yàn)數(shù)據(jù)為基礎(chǔ)構(gòu)建了海圖源數(shù)據(jù)庫原型；對(duì)ENC產(chǎn)品規(guī)范中關(guān)于數(shù)據(jù)內(nèi)容的規(guī)定進(jìn)行了總結(jié)；以ENC為例，給出了從海圖源數(shù)據(jù)庫中生產(chǎn)ENC新數(shù)據(jù)和更新數(shù)據(jù)的基本示例，初步驗(yàn)證了同源多尺度生產(chǎn)的可行性。
[Abstract]:Digitalization has solved the problem of data from scratch, and information needs to further realize the integration and sharing of data in order to achieve consistency, real-time and high efficiency of data. Although the production of chart in China has been digitized, but compared to the level of marine chart production in developed countries, there is inconsistent product data and update week. In view of these problems, a new chart production system is proposed and its key technologies are studied in this paper. The main work and achievements of this paper are as follows:
1. the domestic and foreign research status of four key problems, such as "sea chart production system", "sea chart data model", "automatic synthesis algorithm" and "spatial data matching", were reviewed and analyzed in the process of production of modern chart.
2. the concept of the homologous multiscale sea chart production system, the framework structure, the main production process and key technology are analyzed. Through the fusion of the multi source chart data, the production system ensures the uniqueness and the potential of the source data and the same origin of the product data. At the same time, there is a derivative relationship between the product data and the cause of avoiding the source. The difficulty in data production is caused by the large scale span of data and product data, and the "spatial geometry instance sharing pool" is introduced to realize the sharing of spatial geometric data between different product data and ensure the consistency of the product data.
3. in view of the redundancy of data in the entity model, it is easy to cause data inconsistency, and the topology model is too complex and redundant. In this paper, a seamless and stepless spatial data fusion model is proposed, which not only can achieve the fusion of multi source spatial data, but also avoids the data redundancy in the traditional spatial data model. In combination with the international standards of S-100 and S-57 two chart data, the sea chart element model and sea chart attribute model are given, and the "minimum topology" space model is proposed. At the same time, a set of sea map source database design and reality scheme is put forward according to the sea map source data fusion model: the sea chart concept dictionary and the element list are put forward. It realizes the unification of the semantics of chart data, realizes seamless seamless expression of the chart source database by decoupling elements set and space set.
4. in view of the problem that the external data is inconsistent with the data of the chart source, this paper proposes a transformation language based on XML, which is composed of pronouns, nouns, adjectives, conditional statements, operation sentences and other grammatical structures. The automatic conversion of elements and attributes is realized, and the automatic transformation of spatial patterns is realized by the topology reconfiguration technology, and then the D is realized. The transformation of external data from NC and CDC data to the sea map source data; the automatic matching method of the elements in the new data and the source database is studied, and an improved Hausdorff distance based on the error theory is given. Finally, the quality evaluation factor and the quality evaluation rule knowledge between the different sea map source data are analyzed and summarized.
5. in order to derive the product data from the source database, this paper focuses on the automatic synthesis method of the water depth of the sea chart. Through the analysis of the integrated method of the existing water depth point group, it is found that the traditional method regards the water depth group as a plane point set and overemphasizes the spatial distribution characteristics, which causes the automatic synthesis of water depth to be difficult to be practiced. The lack of automatic synthesis method of water depth point group, put forward a new idea of "taking the water depth group as the space point set, taking the seabed terrain feature recognition as the basis for the automatic background water depth synthesis". By establishing the Delaunay triangulation network and the local dynamic updating technology, "the minimum discrete curvature", "the average spacing constraint", "the depth change about" "Bundle" and its combination are constraint conditions, and the gradual automatic generalization of water depth points based on TIN surface is realized.
6. in order to update the sea chart source database, we need to transfer the change information to the product data. This paper focuses on the analysis of the updating method of the curve in the product data. Using the curve to constrain the Delaunay triangulation network to establish the CDT-Btree, and through the pseudo bending rejection, resampling and other optimization as well as the parallel bending identification and adjustment and other methods, A curve bending characteristic CDT-Mtree is consistent with visual perception. Based on CDT-Mtree, the curve matching and change recognition of the source curve and product curve are realized by multistage bending matching, and the incremental updating of the change curve in the product curve is realized.
7. gives an experiment on the production of a homologous multiscale chart. Based on the DNC and CDC experimental data, the prototype of the sea chart source database is built. The provisions of the data content in the ENC product specification are summarized. A basic example of the production of the new ENC data and the updated data from the chart source database is given with the example of ENC, and the preliminary verification is made. The feasibility of homologous multiscale production.
【學(xué)位授予單位】：解放軍信息工程大學(xué)
【學(xué)位級(jí)別】：博士
【學(xué)位授予年份】：2013
【分類號(hào)】：P285.7;U675.81

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 錢海忠,武芳,鄧紅艷;基于CIRCLE特征變換的點(diǎn)群選取算法[J];測(cè)繪科學(xué);2005年03期

2 翟京生,陸毅;數(shù)字海圖線性特征的識(shí)別、量測(cè)與綜合[J];測(cè)繪學(xué)報(bào);2000年03期

3 艾廷華,劉耀林;保持空間分布特征的群點(diǎn)化簡(jiǎn)方法[J];測(cè)繪學(xué)報(bào);2002年02期

4 劉穎,翟京生,陸毅,王明濤;數(shù)字海圖水深注記的自動(dòng)綜合研究[J];測(cè)繪學(xué)報(bào);2005年02期

5 艾廷華;;DELAUNAY三角網(wǎng)支持下的空間場(chǎng)表達(dá)[J];測(cè)繪學(xué)報(bào);2006年01期

6 翟仁健;武芳;朱麗;朱強(qiáng);;曲線形態(tài)的結(jié)構(gòu)化表達(dá)[J];測(cè)繪學(xué)報(bào);2009年02期

7 武芳,錢海忠;自動(dòng)綜合算子分析及算法庫的建立[J];測(cè)繪學(xué)院學(xué)報(bào);2002年01期

8 蔣遂平,周明天,戴穎;基于法向的網(wǎng)格簡(jiǎn)化[J];計(jì)算機(jī)學(xué)報(bào);1999年10期

9 劉曉利,劉則毅,高鵬東,彭翔;基于尖特征度的邊折疊簡(jiǎn)化算法[J];軟件學(xué)報(bào);2005年05期

10 應(yīng)申;李霖;劉萬增;王紅;;版本數(shù)據(jù)庫中基于目標(biāo)匹配的變化信息提取與數(shù)據(jù)更新[J];武漢大學(xué)學(xué)報(bào)(信息科學(xué)版);2009年06期

本文編號(hào)：2026089