| DC 欄位 |
值 |
語言 |
| DC.contributor | 土木工程學系 | zh_TW |
| DC.creator | 王中昂 | zh_TW |
| DC.creator | Chung-ang Wang | en_US |
| dc.date.accessioned | 2010-5-30T07:39:07Z | |
| dc.date.available | 2010-5-30T07:39:07Z | |
| dc.date.issued | 2010 | |
| dc.identifier.uri | http://ir.lib.ncu.edu.tw:88/thesis/view_etd.asp?URN=963202091 | |
| 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 | 衛星影像常因許多因素造成對比度不佳的情況，因此常使用影像增揚的技巧來提升對比度，但由於原始影像中較亮或較暗的區域在增揚後會遭到壓縮，造成這些區域的細節資訊流失，無法以人眼判釋。
為解決此種問題，本研究提出以模糊集合理論將單一像元視為數種類別的組合，以混合的程度來得到各類別在像元中所佔之比例，以此種概念來補償資訊遭壓縮而損失的區域。共有三步驟：1. 首先以群聚分類法(Fuzzy c-Means, FCM)對影像進行模糊式分類，以歸屬值表示各類別在各像元的比例。2. 依各類別之歸屬值建立各類別之增揚模型。3. 以歸屬值為權重對各類別之增揚模型所得之增揚後灰階進行重組，求得最後增揚影像之灰階值。得到增揚影像後，以量化評估指標對成果進行評估，包含評斷對比度的Michelson index以及評斷資訊量的Shannon Entropy兩種指標，並與傳統非線性直方圖等化之成果進行比較。成果顯示，本研究之演算法可補償原本可能遭壓縮而失去資訊的區域，增加其資訊量並提供更完整的增揚影像，並且保存傳統方法所得之對比度。
| zh_TW |
| dc.description.abstract | The contrast of satellite images are usually affected by lots of factors. In order to increase the contrast, image enhancement techniques are the easiest methods. However, the darker and brighter area of original image could be compressed and lose detail information. Then user cannot see the detail of these area.
In this study, we provide a fuzzy-based image enhancement method to compensate the brightness lost of the darker and brighter area of the image. There are three stages of the algorithm: First, classify the image by Fuzzy c-Means clustering method. Then we can get the membership value of each class of each pixel. Second, create enhancement model which is based on membership value of each class. Third, set membership value as the right and calculate the gray value of enhanced image. After getting the enhanced image, we evaluate the contrast by Michelson index and the quantity of information by Shannon entropy. Then compare the result and data with the traditional enhancement method. The result indicate that the proposed method could compensate the brighter and darker area and also provide an enhanced image with the same contrast as the traditional enhancement method.
| en_US |
| DC.subject | 影像增揚 | zh_TW |
| DC.subject | 模糊集合理論 | zh_TW |
| DC.subject | 模糊式分類 | zh_TW |
| DC.subject | Shannon entropy | en_US |
| DC.subject | Fuzzy c-Means clustering | en_US |
| DC.subject | Michelson index | en_US |
| DC.title | 以模糊集合理論為基礎之影像對比增揚法 | zh_TW |
| dc.language.iso | zh-TW | zh-TW |
| DC.title | An Image Contrast Enhancement Method bases on Fuzzy Set Theory | en_US |
| DC.type | 博碩士論文 | zh_TW |
| DC.type | thesis | en_US |
| DC.publisher | National Central University | en_US |