本研究主要目的係利用不同時間之SPOT衛星多光譜影像資料，以常態化差異植生指標(NDVI)配合影像相減怯進行汐止地區土地利用變遷偵測之研究。早期的土地利用調查多採用現地探勘或航照判釋等方法，但欲獲取大面積之地表分佈資料，須耗費較長時問與大量經費：隨著科技的進步，使用遙測影像與地理資訊系統技術，經由適當之統計、分析，便可萃取出地物之幾何資訊、空間關係，進而達到快速分辨地表土地利用情形，對於面積廣大或是交通不便的區域，亦可做到全面性之分析與監督，以作爲後續決策參考之依據。因此，本研究將兩時期之衛星影像，作常態化差異植生指標分析後，再進行變遷偵測以判釋研究區內之土地利用變化情形。結果顯示大多數土地利用變遷，係導因於大型社區與都市開發案。
This study is to detect the land use and land cover changes in Hsichih Area by using satellite images techniques. The traditional application of land use/land cover investigation generally depended on great manpower in ground survey that was considered a time and economics consuming work. However, the fast progress of remote sensing technology brought the land use/land cover change detection to a new era. In this study, two different periods of MSS SPOT satellite images were used for land use/land cover change detection of Hsichih Area in Northern Taiwan. The image analytical approaches, including the normalized difference vegetation index (NDVI) and image differencing were applied to highlight the most significant (visible) changing areas. The historical development background and the land use/land cover thematic maps of Hsichih Area were also used as the reference. The results show that the major urbanization projects, such as the constructions of residential communities and the highways, may be the key impact factors of those severely changed areas.
