Who is producing Orthos from the Sony a5100?

The m300 and R360 will be my first RTK setup paired with the emlid rs2+ / rover. My question is geared towards the Sony a5100 accuracy and ortho results and for the sake of the questions let’s say since I don’t have a dji RGB camera for the m300 yet, can I produce cm accuracy ortho’s? The reason I ask is I have seen a few tutorials where rgb does not align with the LiDAR data, which I figure is because the lack of overlap. For missions that having a detailed and accurate ortho is important, I would like to plan it accordingly using the R360. Not ideal, but it will be all I have to start with. To clarify:

  • When exporting the usb from the LiDAR unit, can I save just the photos to my device
  • When paired with the base station during flight, does the metadata of the rgb photos have the calculated coordinates with cm accuracy

The RGB data does not align “perfect” with the point cloud, as the rgb is used to colorize the hundreds of millions of points your lidar produced. Rock even says, during their tutorials, that the “intensity” view is the one you must rely on.

Now, I’ve made many orthos from my Sony camera. The ortho, on its own merits, will suffer from poor photogrammetry practices. You should take the time for proper overlap etc. to make really good orthos. But what about accuracy? Is it centimeter level?

Well, yes. And what I measure is very true to the real world. The problem I have, while the x,y is very good, my z is a function of the ellipsoidal values for height. Therefore, once processed, the “z” will need to be adjusted, just like the Lidar data.

The data from the base station is used, in post processing, to correct the meta data on the photos before processing. This is what gives the imagery the centimeter level precision.
Your photos are part of the data package and will be available once you unpack your data from the Lidar. From there, you can do whatever you want with them.

Generous use of GCPs will help you with calibrations and keeping your data really tight. I would suggest you have an honest discussion with your end user… you might want to call in proper surveyors to tag/measure any important infrastructure/objects/areas that will be subject to intense scrutiny and save the lidar/ortho data as background. Photo ortho data is okay, Lidar is way better… and both are rock solid when combined with solid survey (GCPs).


EXTREMELY helpful, thank you! I expect to have a few more questions once I get started. The rest of my gear arrives today and hopefully will begin training the next few days.

1 Like

I did it in a couple of jobs with sucessful but you have to be carefull with the sidelap and frontlap. the sidelap you can set it in the flight plan but the frontlap you must change the interval of the photos in the R2A/R360 sensor configuration. also you must add more GCPs to get a better solution in the Photogrammetry

1 Like

I have had good luck with generating orthos for traditional sites but less luck with corridor flights. We will probably use a P1 in a separate flight pattern if we need really good ortho imagery. GCPs are essential!

Thanks, Sam. I have been having great luck lately with the sony camera attached to the R360, except when there are many structures, and the edges get a little jagged, but nothing crazy. I don’t have the P1 yet, but I have been dreaming of it and have processed the data before; it’s beautiful!

Yeah it all depends on what the client needs for quality! A lot of my clients forget that good enough is usually more than they ever had to work with previously. Its a fine line to walk for sure!

1 Like