網際網路地圖服務之點位坐標及距離量測精確度評估

Accuracy Assessment on Coordinates and Distance Measurements on Web Mapping Services

Google 於2005年推出谷歌地圖（Google Maps）後，成功將網際網路地圖服務（Web Mapping Services, WMS）推廣至一般使用者，成為目前一般使用者最直覺獲取空間資訊的方式。但美國國家地理空間情報署（National Geospatial-intelligence Agency, NGA）以及國際油氣生產者協會（International Association of Oil & Gas Producers, IOGP）卻相繼建議使用者勿在網際網路麥卡托投影（Web Mercator Projection）上進行重要量測工作。在網際網路地圖上量測，除了會受限於原本圖資的測製精度之外，也會受到地圖投影方式、螢幕點選精度、圖台設計等影響，導致量測結果與實際有所出入。本研究以使用者觀點出發，採用實驗之方式，試圖釐清網際網路地圖量測精確度，以提供一般使用者參考。本研究以Google Earth、ArcGIS Earth 以及內政部國土測繪中心臺灣通用電子地圖臺灣通用電子地圖為實驗圖台，以臺灣本島地區119°E、120°E、121°E三條經線及23°N、24°N、25°N三條緯線所交會出的 9個整數經緯度交點為坐標量測目標，以兩點之間的12條邊長作為距離量測標的，評估使用者在使用上述網際網路地圖服務時所能得到之精確度。 Since Google launched their first open access web map application, Google Maps, in 2005, the web mapping service(WMS) has been widely used by people all over the world. The convenience and open accessibility of the WMS breaks down the high barrier of using professional paper map. Nowadays, people has been used to measuring coordinates and distance on the WMS rather than on the paper maps now. However, the U.S. National Geospatial-Intelligence Agency(NGA) and the International Association Oil and Gas Producers(IOGP) successively publish notices to suggest people do not measure distance on the WMS based on the Web Mercator Projection. Distortions are inevitable for any projection, therefore, every map contains distortion. But the measuring error on the WMS derives not only from the projective distortion but also from the human-machine interface and the web map platform. The objective of this research is to assess the overall accuracy of the measurement on the WMS. We choose the 9 intersection of 3 meridians, 23°N, 24°N, 25°N, and 3 parallels, 120°E, 121°E, 122°E, as the point target. The distances of the 12 edges formed by the 9 intersection are chosen for the accuracy assessment on the distance measurement. Three web mapping services, the Google Earth, the ESRI ArcGIS Earth, and the NLSC Taiwan Electronic Map are selected for the experiments. First, the Web Mercator Projection is reviewed and its projection formula is used to assess the theoretical accuracy of the 9 intersection points’ coordinates. Second, we invite several participants to click 10 time on each of the 9 intersection points to evaluate the experimental accuracy on the coordinates. Finally, the measured distance of the 12 edges are compared to the geodesic length to evaluate the experimental accuracy on the distance measurements. The final results of this research can be served as a reference while people want to make measurements on the WMS.

Since Google launched their first open access web map application, Google Maps, in 2005, the web mapping service(WMS) has been widely used by people all over the world. The convenience and open accessibility of the WMS breaks down the high barrier of using professional paper map. Nowadays, people has been used to measuring coordinates and distance on the WMS rather than on the paper maps now. However, the U.S. National Geospatial-Intelligence Agency(NGA) and the International Association Oil and Gas Producers(IOGP) successively publish notices to suggest people do not measure distance on the WMS based on the Web Mercator Projection. Distortions are inevitable for any projection, therefore, every map contains distortion. But the measuring error on the WMS derives not only from the projective distortion but also from the human-machine interface and the web map platform. The objective of this research is to assess the overall accuracy of the measurement on the WMS. We choose the 9 intersection of 3 meridians, 23°N, 24°N, 25°N, and 3 parallels, 120°E, 121°E, 122°E, as the point target. The distances of the 12 edges formed by the 9 intersection are chosen for the accuracy assessment on the distance measurement. Three web mapping services, the Google Earth, the ESRI ArcGIS Earth, and the NLSC Taiwan Electronic Map are selected for the experiments. First, the Web Mercator Projection is reviewed and its projection formula is used to assess the theoretical accuracy of the 9 intersection points’ coordinates. Second, we invite several participants to click 10 time on each of the 9 intersection points to evaluate the experimental accuracy on the coordinates. Finally, the measured distance of the 12 edges are compared to the geodesic length to evaluate the experimental accuracy on the distance measurements. The final results of this research can be served as a reference while people want to make measurements on the WMS.