But then we must preprocess the data before load it in QGIS

>________________________________
> De: Randal Hale <rjh...@northrivergeographic.com>
>Para: qgis-developer@lists.osgeo.org 
>Enviado: Miércoles 29 de enero de 2014 14:49
>Asunto: Re: [Qgis-developer] Lidar data into QGIS
> 
>
>Would LASTools add anything to the conversation? I do believe they 
>re-order the file (but I've not used it).
>
>
>-----------------
>Randal Hale, GISP
>North River Geographic Systems, Inc
>http://www.northrivergeographic.com
>423.653.3611 rjh...@northrivergeographic.com
><mailto:rjh...@northrivergeographic.com>
>twitter:rjhale
>http://about.me/rjhale
>
>On 01/29/2014 08:25 AM, Jean-Roc Morreale wrote:
>> Most lidar files are not spatially ordered so with a sequential read 
>> you will end up reading the whole file, it's like accessing a untiled 
>> unpyramided bigtiff : painful.
>>
>> Le 2014-01-29 13:52, A Huarte a écrit :
>>> I think but using a quadtree approach there is a problem, and it is
>>> none other than the construction of the quadtree itself for so big
>>> data.
>>>
>>> IMHO I would use a specific provider (with a specific featureiterator)
>>> reading sequentially the LAS file and discarding points before fetch
>>> it to renderer using the RectangleFilter of current FeatureRequest. It
>>> avoid create many-many QgsFeature's that after are discarded in
>>> renderer side.
>>>
>>> I use this technique in a personal viewer, and after first read of
>>> file (the file is cached in SO), the following drawings are quite
>>> fast.
>>>
>>> Alvaro
>>>
>>>> -------------------------
>>>> DE: Jean-Roc Morreale <jr.morre...@enoreth.net>
>>>> PARA: qgis-developer@lists.osgeo.org
>>>> ENVIADO: Miércoles 29 de enero de 2014 13:32
>>>> ASUNTO: Re: [Qgis-developer] Lidar data into QGIS
>>>>
>>>> Can it performs fast enough to display hundred millions points or
>>>> billions at once ? Does it leverage a quadtree approach ?
>>>>
>>>> Le 2014-01-29 12:55, A Huarte a écrit :
>>>>> yes, it can be implemented on renderer level (Martin Dobias
>>>> propose
>>>>> it) or on featureiterator level similar to current simplification
>>>>> classes for line and polygons
>>>>>
>>>>> I think it is pending for a test of performance
>>>>> Best Regards
>>>>> Alvaro
>>>>
>>>> _______________________________________________
>>>> Qgis-developer mailing list
>>>> Qgis-developer@lists.osgeo.org
>>>> http://lists.osgeo.org/mailman/listinfo/qgis-developer [1]
>>>
>>>
>>>
>>> Links:
>>> ------
>>> [1] http://lists.osgeo.org/mailman/listinfo/qgis-developer
>> _______________________________________________
>> Qgis-developer mailing list
>> Qgis-developer@lists.osgeo.org
>> http://lists.osgeo.org/mailman/listinfo/qgis-developer
>
>
>_______________________________________________
>Qgis-developer mailing list
>Qgis-developer@lists.osgeo.org
>http://lists.osgeo.org/mailman/listinfo/qgis-developer
>
>
_______________________________________________
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

Reply via email to