How do I deal with spurious events that cause poor RGB mosaicking (APX-15 systems)?

If the number of events compared to number of images is small, this may be an easy fix. These extra events typically come at the very beginning of very end of the flight. If so, follow option 1. If the number of events is much greater than the number of images, this may be a more involved process. Follow option 2. 

1. Small amount of extra events
   1. Backup event1.dat in processed APX folder
   2. Open event1.dat in processed APX folder 
   3. Remove the extra number of events from the beginning of this file 
   4. Re-process 
   5. If this doesn't work to fix the mosaic, revert to the original event1.dat file. For c. instead of removing from the beginning of the file, remove from the end of the file. 
2. Large amount of extra events
   1. Excel method to determine if events are spurious (extra events captured at nearly same time as another event). 
      1. Open event1.dat in Excel 
      2. Highlight first column (timestamp) and set conditional formatting to highlight any events with timestamps less than 1.4 (seconds). 
      3. You may find that the number of these short timestamp delta events matches up with the number of extra events. If so, remove these events, relabel the event # column numerically, save, and try reprocessing. 
   2. Manually checking images compared to DSM or VNIR mosaic in QGIS 
      1. Open the GPRO\APX-15 events file in QGIS. 
      2. Pull in the VNIR mosaic if available, or DSM if not. 
      3. Pull in the GPRO\APX-15\trajectory.json file. 
         1. This will import two items, the trajectory line, and the event points. 
      4. Open symbology for the trajectory line. 
         1. Change line type from Simple to Interpolated. 
         2. Change color from Single to Varying. 
         3. Set Start Value to 0, End Value to 1. 
         4. Apply. This makes it easy to see the start and endpoints of the trajectory, and follow along with the direction of travel. 
      5. Open the RGB images in a photo viewer.  
      6. Run through the images manually in photo viewer.  
         1. Compare each image with the event point along the trajectory and the VNIR mosaic or DSM imagery below it.  
         2. See if you can determine which images matches with each event.  
         3. If any events exist that do not have images, remove them from the event1.dat file. If any images exist that do not have an event, remove the image.