| DC 欄位 |
值 |
語言 |
| DC.contributor | 土木工程學系 | zh_TW |
| DC.creator | 周明中 | zh_TW |
| DC.creator | Min-Chun Chou | en_US |
| dc.date.accessioned | 2005-7-14T07:39:07Z | |
| dc.date.available | 2005-7-14T07:39:07Z | |
| dc.date.issued | 2005 | |
| dc.identifier.uri | http://ir.lib.ncu.edu.tw:88/thesis/view_etd.asp?URN=92322091 | |
| dc.contributor.department | 土木工程學系 | zh_TW |
| DC.description | 國立中央大學 | zh_TW |
| DC.description | National Central University | en_US |
| dc.description.abstract | 外來植物在人為刻意的大量栽植與繁殖以求在短期內獲取經濟利益的情況下，使得其數量大量增加，又未加以適當管理，進而造成對原有生態入侵之入侵樹種，因此對於外來植物物種的監測是當前的重要課題。目前對於外來植物的監測主要是採用人工現地調查的方式，不但費時耗力，而且對於調查成果掌握不易。隨著遙感探測技術的進步，應用遙測影像自動化分類便成為許多使用者之興趣所在。因此若能利用遙測技術在外來植物的監測上，會更省時省力且較為經濟。
現今對於遙測影像的分析方法可分為光譜或紋理兩種方式，由於高解析力影像含有豐富的紋理資訊，尤其有利於紋理分析。本研究之目的在於利用GLCM紋理分析法進行高解析力衛星影像的分析以偵測恆春地區銀合歡的分布狀況，由於GLCM分析時移動視窗大小之選擇對於紋理分析之結果影響相當大，因此本研究先使用半變異元圖分析出較適合分類銀合歡及其他種植物之移動視窗大小。接著加入其他不同之GLCM分析參數如像元對方向及量化方式，計算出多張紋理影像。然後針對此多張紋理影像進行PCA轉換，選取最具差異性的紋理特徵加入原始光譜資訊，再使用最大似然法進行分類，最後使用地真資料進行檢核。 | zh_TW |
| dc.description.abstract | The rapid spread of nonnative plant species have caused considerable negative impact to the biodiversity and ecosystems in Taiwan due to man-made planting in order to gain economic profits in short-term. For example of Leucaena, it has become a dominant species in south Taiwan and has caused an inbalance in local plant community. To address this issue, it is necessary to obtain correct information about the current status of nonnative species.
Currently, the primary method of mapping invasive plants is field investigation, which is not only time-consuming but also very expensive. On the other hand, remote sensing offers a more timely and economical alternative. As the quality and availability of high resolution satellite imagery has improved dramatically in recent years, the analysis of high resolution images has become an important method in a varity of remote sensing researches and applications.
The purpose of this study is to apply texture analysis of high resolution images to identify texture features that can distinguish target plants from other objects. The texture features can then be used to classify high resolution images in order to monitor the invasive plant species (Leucaena). Test results indicate that texture features obtained from this research can be used to successfully map Leucaena in the Ken-Ding National Park of southern Taiwan using high resolution satellite images. | en_US |
| DC.subject | 紋理 | zh_TW |
| DC.subject | 入侵性植物 | zh_TW |
| DC.subject | 分類 | zh_TW |
| DC.subject | invasive plants | en_US |
| DC.subject | classification | en_US |
| DC.subject | texture | en_US |
| DC.title | 紋理輔助高解析度衛星影像分析應用於偵測入侵性植物分布之研究 | zh_TW |
| dc.language.iso | zh-TW | zh-TW |
| DC.type | 博碩士論文 | zh_TW |
| DC.type | thesis | en_US |
| DC.publisher | National Central University | en_US |