本研究使用台灣彰化及雲林兩個地區之2009年~2012年的歷史衛星影像

本研究使用台灣彰化及雲林兩個地區之2009年~2012年的歷史衛星影像，配合2012年春夏及秋冬兩季之現地調查驗證作業，進行土地利用類別之監督式自動化判釋，藉由年度及分類之加總來獲取農業總耕地面積，至於該等時間序數據再藉由灰色建模來估測2013年的總耕地面積，此一估測值與SPOT-6遙測影像的自動判釋成果進行比較，誤差在合理範圍內。本文提供了一個獲取當期或當年度之總耕地面積的快速方法，而透過灰色模式亦能快速估測下一期或接續年度的總耕地面積，該等資訊可提供農政單位管理上的參考。

Results (English) 1:

This research using Taiwan Changhua and the Yunlin two a area of 2009 years ~2012 years of history satellite image, tie 2012 spring summer and the fall/winter two quarter of field survey validation job, for land using category of supervision type automation sentenced release, through annual and the classification of added total to gets agricultural total arable land, as the, time sequence data again through grey built die to estimates 2013 years of total arable land, this a estimates value and SPOT-6 telemetry image of automatically sentenced release results for compared, errors in reasonable range within. This article provides one gets when or when the annual quick method of total arable land, and through the grey pattern can quickly estimate the next successive annual or total cultivated area, which provides agricultural authorities on the management of information such as reference.

Results (English) 2:

In this study, using historical satellite images of Taiwan's Changhua and Yunlin both regions from 2009 to 2012, with the current investigation to verify the job in 2012 spring and summer and autumn and winter, the land-use category of supervised release sentence automation, with annual and increase the general classification of the total agricultural land area to obtain, as these time series data and then by gray modeling to estimate the total cultivated area in 2013, automatically determines the outcome of this release the estimated value and SPOT-6 remote sensing images comparison, the error is within a reasonable range. This article provides a quick way to get the current period or when the total cultivated area in the year, and through the gray mode can also quickly estimate the total cultivated area or follow the next year, such information can provide a reference agricultural authorities management.

Results (English) 3:

This study used two regions of Taiwan's Changhua and Yunlin 2009 to 2012 years history of satellite images, with 2012 Summer and winter seasons in situ investigation and verification work, land use categories of supervised automated interpretation, by year and classification of added to obtain the total land area of agriculture, as for the time series data again by grey modeling to estimate 2013 total arable land area, the estimate of the value of the SPOT-6 remote sensing images automatic interpretation results are compared, error in a reasonable range. In this paper, we give a access to current or when a fast method for annual total cultivated area.And through the grey model can fast estimate of next period or subsequent annual total arable land area, the information may provide as a reference for the agricultural unit management.