Terrain with cracks

We are using Cesium Terrain Builder (https://github.com/geo-data/cesium-terrain-builder) to tiling our DEM data (.tif files) in to cesium height map terrain format, while the out comes is not good, especially at a close glance. First, there are cracks on the globe,seems that the skirts between tiles are not working appropriate. Sceond, the surface is not smooth but zig-zag like. We guess it may caused by some issues of cesium terrain rendering. What should we do to fix this issue?

Hello,
for information, I have the same issue with GeoserverTerrainProvider (see here) which uses also height map terrain format. This issue was identified in january 2016 by GeoserverTerrainProvider users (https://github.com/kaktus40/Cesium-GeoserverTerrainProvider/issues/14).

Great,very helpful information. Since the problems of serrated surface have been solved, I am wondering about the solutions for cracks avoiding. And also I’d ready to try Geoserver Terrain Plugin. Thanks !
在 2017年5月21日星期日 UTC+8下午1:32:55,farouk abdou写道:

I have tried Geoserver Terrain Provider with BIL format while there is no terrain loaded (the globe is dark with nothing), I traced the resources loaded in browser console, while the only log out information is Object {heightMapWidth: 65, heightMapHeight: 65, formatImage: Object, formatArray: Object, tilingScheme: undefined…} and nothing else, the network resources seems only loaded the xml information of my WMS. The version of my Cesium tested is 1.26 and the code is :
var terrainProvider = new Cesium.GeoserverTerrainProvider({

url : “http://localhost:8003/geoserver/DEM/wms”,

layerName: “DEM:globe”,

styleName:’’

});

the Corss region polices have been solved, is there anything I missed ?

在 2017年5月21日星期日 UTC+8下午1:32:55,farouk abdou写道:

Hello,
from what I see, tilingScheme = undefined means that Geoserver Terrain Provider was unable to defined your CRS or SRS from your WMS capabilities document (xml document loaded from geoserver). The CRS identified by the plugin are detailled here in the code. You should check if you have a match with your WMS capabilities.

Thanks,the SRS of my data is right. The error is caused by the ‘layerName’ I used which should be ‘globe’ not ‘DEM:globe’. I noticed that layerName should be the layer name in workspace while here I used the workspace DEM, after I removed the workspace tag, it works. However, the terrain is not right,it is far from the surface of it in arcglobe. For excample, the playground is flat in arcglobe while is a slop in cesium.

在 2017年5月23日星期二 UTC+8下午12:49:34,farouk abdou写道:

If there is an issue that I found is before I set the parameter of maxLevel the surface of globe is not right (figure 1), after the parameter seted is appears right (figure 2).

(figure 1)

(figure 2)

在 2017年5月23日星期二 UTC+8下午2:19:56,lmw…@gmail.com写道:

It’s not an issue. By default the maxLevel is setting to 11. It seems you have a very high detailed terrain. Maybe, you should set maxLevel at least at 15?

Yes, I set the level at 16 (almost near the resolution of my data) and worked well with far view point while the surface still have cracks and appears serrated at close.
在 2017年5月23日星期二 UTC+8下午4:21:22,farouk abdou写道:

i have a question on , converted the tif into terrain.
But, sir I am still stuck with .tif file.

I have tried my luck with Cesium Ion, but was unable to get the exact terrain location, and end up with the whole globe.

So, sir, it would be very kind act on behalf of you, if you could kindly enlighten me with the method, using which you were able to convert your .tif file into .terrain file.(so that I could be able to see all the elevations and dimenstions of the features of the earth surface)

I am in great need of this help.

Hope you will understand.

Thank You

Hi there,

When you process terrain data in Cesium ion, you have to generate a whole globe data set in order to display it in Cesium. You will have your higher resolution terrain data in one area, and the rest of the globe will be a base lower resolution. You can use a tool like QGIS to find the location of the area of interest to view your higher res data.

Thanks,

Gabby