Hi, I have 2 questions
I'm using ArcPro 2.7.1
1) I'm trying to add in images from a drone, DJI Phantom 4 pro. When I use a new workspace I am able to see the images and they recognize the spatial data.
However, if I add the images individually the spatial data is not recognized. The images are in jpg format. Having the same issue with drone videos as well.
2) I have a lot of stacked images and I need to do some calculations before they are orthomapped. How do you basically "select by location" for the rasters before they go to raster calculator. I am making a model that will flow like this 1)images come in 2) images are sorted by location 3) raster calculator 4) orthomapping.
Thanks so much for your help
I should be able to help you, but I may need to clarify my understanding of what you're doing. From what you said, I think this advice should help:
First, note that geospatial video is processed differently than single JPG images. If the video is captured by a DJI drone using the free flight app available for iPad, it will simultaneously record the required metadata to process and view the video using the Full Motion Video capability of ArcGIS Pro (with Spatial Analyst). Unfortunately, video captured without using that iPad app will not have enough metadata about both drone location (XYZ from GPS) and the orientation of the video camera (heading, pitch & lens field of view) to work in ArcGIS.
Single JPG images from a drone are referred to as "geotagged" but that GPS information is not enough to bring the images directly into ArcGIS Pro. If you have an advanced license, and you create an ortho mapping workspace (specify type=drone), ArcGIS Pro can apply photogrammetric processing to generate an orthomosaic of your images. This video should be helpful.
In your question you say "When I use a new workspace I am able to see the images" - If you were not referring to the ortho mapping workspace, please let me know.
When you say "I add the images individually" I'm assuming you mean you simply drag/drop single images into the map? You're correct, that will not work with simple geotagged images - they need to be processed first, using photogrammetry, along with many other overlapping images, to achieve proper georeferencing to create an orthomosaic. If you are seeking to use the individual photos after processing (vs. the orthomosaic of many images), let me know and I can explain how to do that.
Last, I'm not sure what you mean by "many stacked images". If you mean many images in the same location, are they already georeferenced, so they are all placed into the correct location on the map? If that is true, you should not need ortho mapping. If they are multiple photos of the same location but not yet georeferenced, ortho mapping may work for you but this would be an unusual use case. Let me know more about your input data and your objectives, and I will do my best to advise you. If you are seeking to use the raster calculator, I believe that should be applied after georeferencing (either using ortho mapping or possibly another method).
Thank you for taking to the time to answer my questions.
Thanks, for the video information. I wasn't using the iPad just an sd card. I'll try with an ipad and see if I get different results.
So, first, the workspace I was referring to was the orthomapping workspace.
I've tried adding just one image to the orthomapping workspace. Or an image stack (which I do mean images all taken at the same location, using burst mode on the drone). The workspace does not accept this. The intent of the workspace is to create a mosaicked dataset and one image or a set of stacked images does not work for that.
Basically, I need to be able to georeference without using the ortho mapping workspace. The images are only geotagged not georeferenced. Any insights?