Anybody used a Zeb Revo handheld scanner

Or the Zeb 1

I did a largish job a few months back with a hired Zeb1. Discussion thread here ( https://surveyorconnect.com/community/threads/internal-building-surveying.327223/ ).

From a message I sent someone else:

Tips:
- Watch all of the Geoslam youtube videos.
- Try to do your survey on a day where there is little or no people in the building. We were lucky in the fact that no one was using that section of the hospital on a Saturday and we spent a bit of time opening up doors (get lots of door stops to hold doors open). The less movement (doors, people, etc) whilst you are scanning the better.
- Ensure that you walk in loops to close the survey.
- Cover up mirrors if possible as the scanner doesn't like them. It will show the data scanned by mirror reflection (essentially the wall opposite the mirror) behind the mirror.
- We initially had a number of the scans go a bit wrong but after emailing the geoslam guys in the UK, they were able to delve into the data and fix the scans.
- We used cloud compare ( http://www.danielgm.net/cc/ ) to extract data. My boss spent quite a bit of time learning how to use that to get some good results out if it. I'm sure there is better software available but cloud compare is free so we used it to learn with.
- I used the total station to measure a number of points around the building, wall corners, ceiling & floor levels, etc) and then imported those as a dwg into cloud compare and registered the scan to the dwg points. Checks with total station and disto at end were in the order of +/-20mm mostly.
- Looking at a slice through a wall you get a bit of noise, ie most points will be accurate and some will be on either side of the wall survey (+/-50mm). We got good results by taking horizontal slices as an image and then digitising them in Terramodel and Global Mapper (which is what we use normally). We tried to avoid digitising but with all of the stuff in the rooms we scanned, it turned out the be the best option.

Overall time wise, we probably spent more than we would have had we done it by traversing around the halls and using a disto for the rest. However the time spent on site was vastly less (which was great for the client) and the rest was done in the office. As we had a steep learning curve, the data processing would be significantly quicker next time.

Hi we tested nearly every scanner when we got into scanning. Zeb,faro,trimble,leica,topcon,reigl,stonex. We tried using photomodeller. Through all of that I can recommend this. The faro120 actualy only has a 40m range effectively. Scannung further than that the scan data is fairly useless. Faro330 has a range of maybe 130m. Option 1 : If you are gonna do <30 scans get the faro. Cheaper to rent and data is fine for interior stuff. Learning curve is not bad and you can register the scans in scene (30 day trail) or recap pro.

Option 2 : more than 30 scans get a leica p20/30/40. More expensive to rent, and cyclone has a fair learning curve but data is much much better. You can recognise the difference in a leica vs faro scan instantly on the noise (time of flight vs phase). The zeb will be fine for maybe a house or few rooms but it doesnt work nice for a whole building. If you get the leica dont register on board. Its slower no matter how much you hope it to be faster. And pictures on the lieca makes is super slow. Just do scans and pyt in cyclone, register and there is a way to export the scan locations into recap to view a panoramic scan.

Sent from my SM-N920C using Tapatalk

Seb, post: 392127, member: 7509 wrote:

Overall time wise, we probably spent more than we would have had we done it by traversing around the halls and using a disto for the rest. However the time spent on site was vastly less (which was great for the client) and the rest was done in the office. As we had a steep learning curve, the data processing would be significantly quicker next time.

I think that's what I am worried about in that the learning curve might be a bit steep so I will have to use it on a job where there is a suitable timescale. Did you bring the cloud directly into CAD for extracting any of the floor plans?

RSAsurv, post: 392155, member: 10950 wrote: If you get the leica dont register on board. Its slower no matter how much you hope it to be faster. And pictures on the lieca makes is super slow. Just do scans and pyt in cyclone, register and there is a way to export the scan locations into recap to view a panoramic scan.

Sent from my SM-N920C using Tapatalk

I am trying to get my head around the workflow for registering. I have used a Topcon scanner where I was able to back-sight to control on each setup but I imagine this will be very slow for larger buildings. Do you use targets for scanning room to room, how many targets does each scan need or do you just register to similar points in each scan. If so how much of an overlap does it require?

Yes on board back sight registering is very slow in the field. On lets say a 1000m2 floor we would have 6 targets per floor. Scans get registered to other scans by only doing 1 floor at a time using visual alignment in cyclone and then gets link to control. Then you go to the next floor. On a building we did we put 6-8 targets per floor on that 1000m2. End of each corridor basically. Almost like cloud to cloud but more manual so you have more control. With leica you have about 50m range where overlap can be used to do alignment. Lets say its a corridor. Then do a scan 5m from the door. 1 in the door. And 1 inside. Scan using level 1 detail at 12mm spacing (gives a 53sec scan time). We do about 200 scans a day like that if its a building inside

Sent from my SM-N920C using Tapatalk

check out this thread.
https://surveyorconnect.com/community/threads/internal-building-surveying.327223/page-3#post-379521

Hey Totalsurv, if interested in viewing the Geoslam range of products contact us at LES. We are distributors for Geoslam in Ireland and would be happy to answer any questions you have.