关于google-earth中的比例尺、图像精度和eye&alt的问题
&谷球，也就是谷歌地球，google
earth，堪称划时代的创新，不是因为它提供了最好的卫星图（显然不是），而是因为
1 其卫星图可以极度方便的免费获得，前所未有的方便。
2 其卫星图的整合界面极为简单，也就是一个地球的形状
有完整的全球信息，虽然有些地方精度不足，但总体细节够多。
可以说，谷球改变了大众对地球的基本视角。必须足够简单方便，影响足够广泛，细节足够精细，才称得上“全人类的基本的”视角。史上第一次，普通人可以真正有效的以全球的、深入细致的角度去观察地球。无论对比传统实体地球仪，还是厚度惊人的地图册，或者昂贵的商业图，谷球显然开创性的做到了这一点。
从此我们观念中的世界图像，将是真正完整的，丰富深入细致立体的，一个非常接近真实地球的虚拟地球。这只有谷球这样对公众开放的海量数据系统才能实现。这东西对于学术研究真是太好了。
最近正在通过谷球调研全国汉地主要佛寺，为了能在统一比例尺下对比研究，我关心了一下谷球的比例和精度问题，有如下结论：
1 首先要记得在view菜单里打开比例尺。
同样比例尺数值的图，全屏和非全屏，显示的内容宽度相同，高度随比例变化，像素随放缩情况变化，这显然是谷球程序的预设。同样相对高度的视点下，屏幕所显示的地图内容的宽度是不变的，无论是否全屏，如何缩放。而比例尺的长度固定等于图像宽度的四分之一。
3&右下角的eye
alt的意思，官方解释为：&The Eye Alt figure at the
bottom of the 3D viewer refers to eye altitude (the elevation of
viewpoint).显然，这一海拔高度是那个虚拟相机的高度（眼球海拔），其视角应当是固定的。需要注意的是，此标高需减去图像所在标高才能得到相对地面的视点高度。通过在海平面测试，眼球海拔/屏幕宽度=0.86左右。由此推得，谷球的相机的视角大概是60度。不出所料，果然是这个经典角度。
平移图像时，比例尺会变化。应该是由于海拔变化引起的。如果从低海拔平移至高海拔，视点不能低于海拔，所以可能会强行升高视点。
5&要微调比例尺，可以用alt键加上数字键盘的+或者-键。
6&如果不严格控制截图比例，依靠比例尺后期缩放也是很方便的。
如果要统一比例尺截图，最好是在定点是设定其range，也就是眼球的相对高度。但是比例尺和屏幕宽度不是此高度的整倍数。此数值无法批量调整（很遗憾，建议改进）。图面比例尺为100M时，range为346米。
地球经纬度，每0度0分1秒的球面距离大概是30米。谷球的精度能到15米，也就是半秒。据说新卫星能到3米，似乎目前已经上天了，详情我不知道。。。
谷球据说有20级精度，放大到最大精度以后，再放大就只是插值显示了。我不能确定谷球原始图的分辨率，但通过比较不同比例尺输出图像，我觉得大概比例尺在100米时，基本上谷球显示的是1：1的图像，此时截图应该效率最高。谷球大概在比例尺300M时就已经在使用最高图像，不过是缩小显示的。我也觉得似乎50米时的截图细节会略丰富一点点，但50米时显然已经是插值放大了。所以也可能是在比例尺80多米，也就是眼球高度300米的时候为1比1显示。当然也可能谷球是用的英制，那就不好说了。总之，谷球1：1显示图像时的比例尺肯定在150-50之间，按照100M截图差不多。注意，此精度和显示器分辨率有关。我的显示器是。输出谷球图宽是1280像素，比例尺100米。其他分辨率的显示器可以类推，也就是说，400米宽的图，像素是1280宽，真实图像精度是
3像素/米。你的显示器像素越多，你的最佳截屏比例尺越长。当然这只是估算。实际谷球卫星图的精度，只有它自己知道。。。另外，谷球软件的输出默认设置似乎对图像精度也有影响，详情未知。
据附后的wiki文档指出，一般全球图的精度是15M/像素，高度的精度最高10M，不过这里要注意，图像的分辨率精度和谷球的矢量定位精度是不同的。具体两者的实际精度，应该有谷球的官方白皮书说明才对，可惜我没有。如果谁找到了欢迎提供。
12&以上为本人瞎猜的，欢迎讨论。转载请注明出处。
这样有讨论就可以反馈给我以便改进。谢谢。
以下wiki的文章：感谢水木敢达版的萝莉控FaYuri同学提供资料并与我讨论。
Resolution and accuracy
The Isles of Scilly, showing the very low resolution of some
islands. The islands (green area) are about 10 km
across.49°56′10.81″N 6°19′22.88″W&
/& 49.9363361, -6.3230222& (Low
resolution Isles of Scilly)
The Isles of Scilly, showing the very low resolution of some
islands. The islands (green area) are about 10 km across. [show
location on an interactive map] 49°56′10.81″N
6°19′22.88″W& /& 49.9363361,
-6.3230222& (Low resolution Isles of Scilly)
The west side of Gibraltar, tilted view showing the sea rising up
the Rock of Gibraltar - claimed altitude of the sea just off the
beach at Elliots Memorial, 252 m. 36°6′59.6″N
5°21′5.2″W& /& 36.116556,
-5.351444& (Water altitude problem in Google
The west side of Gibraltar, tilted view showing the sea rising up
the Rock of Gibraltar - claimed altitude of the sea just off the
beach at Elliots Memorial, 252 m. [show location on an interactive
map] 36°6′59.6″N 5°21′5.2″W&
/& 36.116556, -5.351444& (Water
altitude problem in Google Earth)
Most land areas are covered in satellite
imagery with a resolution of about 15 m per pixel. Some population
centers are also covered by aircraft imagery (orthophotography)
with several pixels per meter. Oceans are covered at a much lower
resolution, as are most notably, Tórshavn,
the capital of the Faroe Islands, and the Isles of Scilly off
southwest England, are at a resolution of about 500 m or less.
These pictures are provided by Terrametrics.
Google has resolved many inaccuracies in the
vector mapping since the original public release of the software,
without requiring an update to the program itself. An example of
this was the absence from Google Earth's map boundaries of the
Nunavut territory in Canada, a territory that had been created on
April 1, 1999; this mistake was corrected by one of the data
updates in early 2006. Recent updates have also increased the
coverage of detailed aerial photography, particularly in certain
areas of western Europe, though not including Ireland where imagery
remains extremely limited.
The images are not all taken at the same time,
but are generally current to within three years. Image sets are
sometimes not correctly stitched together. Updates to the
photographic database can occasionally be noticed when drastic
changes take place in the appearance of the landscape, like for
example Google Earth's incomplete updates of New Orleans following
Hurricane Katrina, or when placemarks appear to shift unexpectedly
across the Earth's surface. Though the placemarks have not in fact
moved, the imagery is composed and stitched differently. Such an
update to London's photography in early 2006 created shifts of
15-20 metres in many areas, noticeable because the resolution is so
Place name and road detail vary greatly from
place to place. They are most accurate in North America and Europe,
but regular mapping updates are improving coverage
elsewhere.
Errors sometimes occur due to the technology
used to measure t for example, tall buildings
in Adelaide cause one part of the city to be rendered as a small
mountain, when it is in fact flat. The height of the Eiffel Tower
creates a similar effect in the rendering of Paris. Also,
elevations below sea level are pr i.e. Salton
City, C Death V and the Dead Sea are all listed as
0 ft when Salton City is approx& 200 Death
Valley is& 286 and the Dead Sea
is& 1,378 ft.
Where no 3 arc second digital elevation data
was available, the three dimensional images covering some areas of
high relief are not at all accurate, but most mountain areas are
now well mapped. The underlying digital elevation model has been
placed 3 arc seconds too far north and up to 3 arc seconds too far
west. This means that some steep mountain ridges incorrectly appear
to have shadows extending over onto their south facing sides. Some
high resolution images have also been misplaced, an example is the
image covering Annapurna, which is misplaced by about 12 arc
seconds. Elevation data was recently updated to 10-meter
(1/3-arc-second) resolution for much of the United States from the
previous 30-meter (1-arc-second) resolution.
The "Measure" function shows that the length
of equator is about 40,030.24 km, giving an error
of& 0.112% compared with the actual value of
40,075.02 km E for the meridional circumference, it shows a
length of about 39,963.13 km, also giving an error
of& 0.112% compared with the actual value of
40,007.86 km.
The Arctic polar ice caps are completely
absent from the current version of Google Earth, as are waves in
the oceans. The geographic North Pole is found hovering over the
Arctic Ocean. There is very low resolution coverage of the
Antarctic continent (1m resolution images of some parts of
Antarctica were added in June 2007 for the first time). The tiling
system produces artifacts near the poles as the tiles become
'infinitely' small and rounding errors accumulate.
Cloud cover and shadows can make it difficult
or impossible to see details in some land areas, including the
shadow side of mountains.
已投稿到：
以上网友发言只代表其个人观点，不代表新浪网的观点或立场。比例尺精度有什么实用价值?_百度知道
比例尺精度有什么实用价值?
提问者采纳
比例尺精度决定了这份地图资料的最高精度，对于这份地图是否满足工作需要的精度要求，首先看比例尺
其他类似问题
比例尺精度的相关知识
等待您来回答
下载知道APP
随时随地咨询
出门在外也不愁大比例尺精度草地资源调查_百度百科
收藏 查看&大比例尺精度草地资源调查本词条缺少信息栏、名片图，补充相关内容使词条更完整，还能快速升级，赶紧来吧！中文名称大比例尺精度草地资源调查英文名称large-scale survey of rangeland resources定　　义对县以下行政区域和基层草地生产单位范围内的草地，进行的成图比例尺大于1：100 000精度，调查成果服务于县以下行政和生产单位的草地生产、建设、利用、保护与规划目的的调查。应用学科资源科技（一级学科），草地资源学（二级学科）以上内容由审定公布
新手上路我有疑问投诉建议参考资料 查看测量员理论考试模拟试题题库
本试题来自：（2010年测量员理论考试模拟试题，）单项选择：二、单项选择题1:5000比例尺的地形图，其比例尺精度为(
)。A．0.05mB．0.1mC．0.2mmD．0.5m正确答案：有， 或者
您可能感兴趣的试题
单项选择题：（)已知等腰直角三角形ABC和等边三角形BDC(如图18-1)，设△ABC的周长为，则△BDC的面积是(
)．A．B．C．12D．E．答案：有，答案解析：有，单项选择题：（)一批图书放在两个书柜中，其中第一柜占55%，若从第一柜中取出15本放入第二柜内，则两书柜的书各占这批图书的50%，这批图书共有(
)．A．300本B．360本C．400本D．460本E．600本答案：有，答案解析：有，
测量员理论考试模拟试题最新试卷
测量员理论考试模拟试题热门试卷何谓地形图?它包括哪些内容?其比例尺精度有什么实际意义?_百度知道
何谓地形图?它包括哪些内容?其比例尺精度有什么实际意义?
大家帮下忙。
我有更好的答案
按默认排序
远比在这里得到的多得多了随便在百度上搜索“地形图”“地形图的内容”“地形图精度”都有很多信息
其他类似问题
比例尺精度的相关知识
等待您来回答
您可能关注的推广回答者：
下载知道APP
随时随地咨询
出门在外也不愁