The Study of the Detection and Examination for the Variation of Cultivated Land Map

• Main Authors: Ying-chiao Liao, 廖英喬
• Other Authors: Rei-yao Wu
• Format: Others
• Language: zh-TW
• Published: 2009
• Online Access: http://ndltd.ncl.edu.tw/handle/99388328275289180617

碩士 === 世新大學 === 資訊管理學研究所(含碩專班) === 97 === The cultivated land is the basis for agricultural management, and is the fundamental data for production survey and yield estimation. Therefore, it is necessary to monitor the land change in a real time basis. However, in-time supports for various applications can not be satisfied, since the cultivated land is distributed all over the island and the overall renewal of the land maps is extremely time-consuming. For the purposes of improving the quality of the cultivated land maps and of providing the comprehensive and up-to-date status of the distribution of the cultivated land, this study establishes the mechanism to assist in monitoring the changes of cultivated land and to refine the checking processes. The objectives of the study are: 1.to examine and organize the cultivated land spatial data 2.to quantify the changes of cultivated land maps 3.to classify the changes of cultivated land maps This study utilizes maps of Taoyuan County as an example for verification. First, identify the spatial characteristics of the cultivated land maps. Confirm the checking items, methods and integration processes adapted consequentially to produce the cultivated land maps which meet GIS specifications and have the characteristics of cultivated land. Then, Compare the data of cultivated land maps with the latest Property Owner’s Data, quantify the variation between section maps by using statistical and inductive method, produce quantitative variation charts and format the charts for usage of field investigation and followed by classification data mining. Finally, Total of 17 classification rules are obtained after the practice. The overall accuracy is 92.94% and each user’s accuracy is above 90%. It shows that only approximately 10% of the cultivated land maps need to be examined manually, which means the improvement of processing efficacy and the decrease of the error rate generated from manual checking.