Couverture WMS

Suite à la mise à disposition la semaine passée de nouvelles orthophotos (Nonfoux, Chexbres et Champ-Pittet) sur notre serveur WMS, une carte permettant de visualiser l’étendue géographique de chaque couche a été créée.

Couverture WMS

Pour rappel, ces orthophotos sont mises à disposition de la communauté OpenStreetMap pour servir de base pour la digitalisation d’objets. Toutes les informations techniques se trouvent sur la page dédiée à notre serveur WMS.

La carte qui est utilisée pour la visualisation des couches est justement basée sur les données OSM. Il est intéressant de voir la différence de contenu entre une zone couverte par une de nos orthophotos et une zone non-couverte.

Différences

Cela nous prouve bel et bien qu’il existe un besoin et un potentiel à mettre des données à disposition pour permettre la digitalisation, c’est une des raisons qui nous pousse à continuer à diffuser nos orthophotos dans ce but.

  1. Hi! Sorry for not writing french, but I use google translate to translate your posts so I can understand them. I have a question:

    The RESOLUTION is 25,20,10cm etc, however what is the ACCURACY of the digitized data? For instance when creating the mosaic of the orthorectified images, which terrain-model is used?

    Keep up the good work, nice to see this nice project going good!

  2. Internal accuracy (i.e. in the bundle block adjustment) is about 2/3 of a pixel. External horizontal accuracy is measured at about 1.5 pixels. Vertical accuracy is about 2x this, i.e. 3 pixels.
    We use either automatically matched DTM or the Swiss official DTM produced by LiDAR a few years ago (MNT-MO). DTM accuracy is not a real issue, since we use the central part of each image for the mosaic. We just make sure that the georeferencing is very consistent with the DTM, or vice-versa.

  3. Hi, thanks for the information. One more question. For instance on the project Orthophoto Yverdon 2007, how many GCPs did you have to measure for this project? It seems that you need many of them?

    Also I wonder if you have done a camera-calibration to get the “lens correction”? I assume yes, but I havent seen any description which method you use to do get these parameters?

  4. Orthophoto Yverdon 2007 was acquired with the R-Pod’s ancestor, named HB-PAX. The number of GCP is not too significant: at least four are needed, eight is comfortable, 12 and more bring reliability and confidence in the block.
    There is no camera calibration as such, but an average radial distortion curve and a self-calibration with 20 parameters. It appears that the camera is of course not “metric”, but parameters vary not much.

  5. I`m interested to know more about the procedures for camera-calibration, the average radial distortion curve and the self-calibration. Can you guide me towards any PDF or articles about this?

  6. I extracted a few pages related to this procedure from the User’s Guide of ORIMA (ORientation MAnagement, the bundle block adjustment software that we currently use)

Laisser un commentaire


NOTE - Vous pouvez utiliser les éléments et attributs HTML tags and attributes:
<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>