查詢結果分析
相關文獻
- 類神經網路與SPOT衛星影像分類之研究
- 類神經網路於多光譜影像分類之應用
- 動態學習法則類神經網路法在闊葉林區土地利用型分類上應用之研究
- 類神經網路應用於遙測影像分類之研究
- A Gary Level Hamming Net for Gray Level Image Classification
- 監督式分類方法於遙測影像判釋之研究
- 以樹狀倒傳遞類神經網路於田埂判釋研究
- Adaptive Filtering Approaches to Multispectral Image Classification
- 非監督式自組織映射圖類神經網路應用於影像分類探討
- Neural Network Procedures for Taguchi's Dynamic Problems
頁籤選單縮合
題 名 | 類神經網路應用於遙測影像分類之研究=Artificial Neural Networks for Classification of Remote Sensing Data |
---|---|
作 者 | 詹進發; | 書刊名 | 國立臺灣大學農學院實驗林研究報告 |
卷 期 | 11:3=217 1997.09[民86.09] |
頁 次 | 頁79-89 |
分類號 | 436.718 |
關鍵詞 | 類神經網路; 影像分類; Artificial neural network; Image classification; |
語 文 | 中文(Chinese) |
中文摘要 | 本研究之目的是在探討類神經網路應用於遙測影像分類之可行性,並利用攝取於 1993 年五月及九月之美國大地衛星影像資料進行分類,以區分十一種地表覆蓋類型。 分類 結果顯示類神經網路可以有效分類遙測影像, 以訓練樣本進行分類之精確度可達 100%,此 外,慎選誤差函數對於分類精確度亦有影響。 |
英文摘要 | The overall objective of this research was to apply artificial neural networks in the classification of remote sensing data. The study area was an agricultural area in northern Colorado. Two Landsat images obtained on May 30 and September 3, 1993, were used in this study. The feed-forward neural network paradigm was implemented to classify 11 land cover types. In addition to the standard sum of squared error function used in feed-forward neural networks, the cross-entropy error function was also used to measure the network error. The results showed that the neural network paradigm was very effective in classifying the data set (e.g., the best classification accuracy obtained was 100% for the training samples). Moreover, the classification results varied when different error functions were used to measure the network errors. |
本系統中英文摘要資訊取自各篇刊載內容。