Laser Scanning Pain Points

[badam]

well yes the best solution is manual cleaning, but you can achieve some result with intensity clipping, maybe color clipping because usualy through window the color will be burn out so it will be 255,255,255 unless hdr is really strong.

i think with ai image recognition and structured cloud we could get some automatic window detection and cleaning, but it would be hard to code. and could be slow.

[Leandre Robitaille]

Do you have any sample datasets you'd be willing to share that include this kind of data, ideally in e57 format with setups.

Hi Shane!

I did a quick scan of my living room. I have a pretty large mirror in the middle of the place that duplicates one side of my house. This type of reflection is exactly what I would label pain points of laser scanning because the reflection fuses with the rest of the data and is hard to clean or even notice.
Did the scan with a S70, scan density at 6.1mm/10m
You can download all files here (this link will be valid for 15 days, pm me if you need a new link);

https://cimao365-my.sharepoint.com/:f:/ ... w?e=toLuzI

I also made a quick Cintoo viewer for people to see the 3D.

https://aec.cintoo.com/837071F361C2B0B2B371

[smacl]

Do you have any sample datasets you'd be willing to share that include this kind of data, ideally in e57 format with setups.

Hi Shane!

I did a quick scan of my living room. I have a pretty large mirror in the middle of the place that duplicates one side of my house. This type of reflection is exactly what I would label pain points of laser scanning because the reflection fuses with the rest of the data and is hard to clean or even notice.
Did the scan with a S70, scan density at 6.1mm/10m
You can download all files here (this link will be valid for 15 days, pm me if you need a new link);

https://cimao365-my.sharepoint.com/:f:/ ... w?e=toLuzI

I also made a quick Cintoo viewer for people to see the 3D.

https://aec.cintoo.com/837071F361C2B0B2B371

Hi Leandre,

Many thanks for the test data set and accompanying cintoo model, exactly what I was looking for. I suspect getting an algorithm to deal with the reflective planar mirrors will be very doable but getting something to handle the partially reflective items like the TV screen will take quite a bit more thought. It will be a while before I get a chance to devote significant time to this but it already has me investigating possibilities.

Shane

[stuartcadge83]

Wow Steve that is a very trippy-looking scan - a conservatory?

Thank you again to everyone who voted in our poll, it's really intriguing to see what bugs people about their current system.

As my colleague Harriet said, everyone who voted in our poll is eligible for 1 week for free with the X7 (UK & Ireland based) so if you fancy being in the running please send Harriet or myself a message with your contact details

Have a good week, All

Stuart
KOREC

[Daniel Wujanz]

LASER SCANNING MARKETING or everything that refers to numbers in a greater sense such as accuracy, data acquisition rates, processing speeds...

The vast majority of numeric information that is exchanged in this industry is meaningless, non-comparable, misleading or just dubious. This certainly attracts clients but also creates a hint of dodgyness. I don't know any other engineering branch were numbers are treated with less respect.

And yes, you can cook with a microwave but there is still a big difference to haute cuisine. Just because a function, feature or sensor shares a similar sounding name does not mean it is the same.

What makes it a pain point is that it takes an awful lot of time to point out differences or explain how something really works instead of creating innovations that (should) drive this industry.

End of rant ; )

Daniel

[RJGEOMATICS]

LASER SCANNING MARKETING or everything that refers to numbers in a greater sense such as accuracy, data acquisition rates, processing speeds...

The vast majority of numeric information that is exchanged in this industry is meaningless, non-comparable, misleading or just dubious. This certainly attracts clients but also creates a hint of dodgyness. I don't know any other engineering branch were numbers are treated with less respect.

And yes, you can cook with a microwave but there is still a big difference to haute cuisine. Just because a function, feature or sensor shares a similar sounding name does not mean it is the same.

What makes it a pain point is that it takes an awful lot of time to point out differences or explain how something really works instead of creating innovations that (should) drive this industry.

End of rant ; )

Daniel

What I really hate is how every sales person out there thinks everyone needs the latest, fastest and greatest.

If your main business is Accurate 3D scans just in Black and white, consisting of only 20 to 30 scans per day, for a few days a week... Guess what. That old C10 is perfectly fine, and does an amazing job on projects such as Floor Flatness and Levelness... Having a faster newer scanner isn't going to save you anything, but you'll get to pay the loan and the sales person will be thankful for their commission.

In general sales people are great magicians at selling you on the fact that, if you get their tool you'll be so much quicker and do more work.... Well guess what, if you are paying staff, it really doesn't matter if they finish up 2 hours earlier in the field, because you have nothing else for them to do in those 2 hours, so they aren't chargeable. And in most metropolis areas the drive time for mobilization is greater than the scan time for progress scanning on items like building construction. So having the latest creates, fanciest and shiniest machine that you are financing, isn't doing anything for your business except putting in debt.

Each business needs do their own assessment. But in any event, no matter what, your pricing and performance needs be based on renting the tool for each job. If you are renting a scanner for 3 days per week consistently, then perhaps it is worth buying one. But buy what you need. you should never justifying buying a tool to get the work. You get the work, and then justify the tools.

Good example is farmers. The one running the old Massey Harris 265 still to this day, probably sleeps on a mattress of money. The farmer who is continually financing the new tractor, is going bankrupt every 5 years and taking advantage of all their clients and creditors.

Make your business and money through blood and sweat. You'll never make a fortune paying the bank and finance companies.

[landmeterbeuckx]

LASER SCANNING MARKETING or everything that refers to numbers in a greater sense such as accuracy, data acquisition rates, processing speeds...

The vast majority of numeric information that is exchanged in this industry is meaningless, non-comparable, misleading or just dubious. This certainly attracts clients but also creates a hint of dodgyness. I don't know any other engineering branch were numbers are treated with less respect.

And yes, you can cook with a microwave but there is still a big difference to haute cuisine. Just because a function, feature or sensor shares a similar sounding name does not mean it is the same.

What makes it a pain point is that it takes an awful lot of time to point out differences or explain how something really works instead of creating innovations that (should) drive this industry.

End of rant ; )

Daniel

What I really hate is how every sales person out there thinks everyone needs the latest, fastest and greatest.

If your main business is Accurate 3D scans just in Black and white, consisting of only 20 to 30 scans per day, for a few days a week... Guess what. That old C10 is perfectly fine, and does an amazing job on projects such as Floor Flatness and Levelness... Having a faster newer scanner isn't going to save you anything, but you'll get to pay the loan and the sales person will be thankful for their commission.

In general sales people are great magicians at selling you on the fact that, if you get their tool you'll be so much quicker and do more work.... Well guess what, if you are paying staff, it really doesn't matter if they finish up 2 hours earlier in the field, because you have nothing else for them to do in those 2 hours, so they aren't chargeable. And in most metropolis areas the drive time for mobilization is greater than the scan time for progress scanning on items like building construction. So having the latest creates, fanciest and shiniest machine that you are financing, isn't doing anything for your business except putting in debt.

Each business needs do their own assessment. But in any event, no matter what, your pricing and performance needs be based on renting the tool for each job. If you are renting a scanner for 3 days per week consistently, then perhaps it is worth buying one. But buy what you need. you should never justifying buying a tool to get the work. You get the work, and then justify the tools.

Good example is farmers. The one running the old Massey Harris 265 still to this day, probably sleeps on a mattress of money. The farmer who is continually financing the new tractor, is going bankrupt every 5 years and taking advantage of all their clients and creditors.

Make your business and money through blood and sweat. You'll never make a fortune paying the bank and finance companies.

I'm doing a lot of construction staking besides scanning. Still using my Sokkia Srx5 which is nearly 11 years old. But it does the job alltough a bit slower then with a new instrument but i'll bet i can do a site faster because i know it flaws but also it strengths. You have to adapt yourself to the hardware of your gear, that's where the money is.

Because i'm solo the gear doesn't get beaten up like in a multi person company so it'll last longer. The strength in my setup is the field software which is on maintenance and has all the new bells and whistles (even ifc import).

[smacl]

In general sales people are great magicians at selling you on the fact that, if you get their tool you'll be so much quicker and do more work.... Well guess what, if you are paying staff, it really doesn't matter if they finish up 2 hours earlier in the field, because you have nothing else for them to do in those 2 hours, so they aren't chargeable. And in most metropolis areas the drive time for mobilization is greater than the scan time for progress scanning on items like building construction. So having the latest creates, fanciest and shiniest machine that you are financing, isn't doing anything for your business except putting in debt.

Great point, getting the logistics right is hugely important for any company involved in survey and scanning. Similarly creating and refining robust workflows around the the hardware and software you're using so that you're extracting maximum efficiency and repeatable quality. As per Lieven's post above, an experienced person working older kit they know will typically have a greater throughput of excellent work than an inexperienced person on the latest and greatest hardware and software. Where the big gains are made in new equipment is in the hands of experienced people in well developed workflows on larger jobs. As we all know, it is very easy to get a fast scanner only to have serious bottlenecks in other places in the workflow that entirely negate the value added by the scanner speed. Also very easy to skimp on the training, where those buying scanning hardware and software don't budget for training and fail to carry out a proper proof of concept exercise before purchasing to see if the total solution is a good fit for them. Many companies who invest in good kit are more reluctant to invest in their people, which leads to a lot of good kit either under utilised or lying idle, inefficient workflows, lost time and plenty of frustration.

[badam]

Other big issue which we are facing is the huge data which should be archived in different state to keep everything intact for future use.

native formats,native project archives (i talk about leica side of things), exported formats (e57, ptx, we never keep this more than a year), formats to the clients (lgs, potree,rcp, etc.. )

with one rtc360 you can generate 20-30 TB of (native) data easily in just one year. And you will need fast access to this. This is huge especially for smaller offices.

[Daniel Wujanz]

Other big issue which we are facing is the huge data which should be archived in different state to keep everything intact for future use.

native formats,native project archives (i talk about leica side of things), exported formats (e57, ptx, we never keep this more than a year), formats to the clients (lgs, potree,rcp, etc.. )

with one rtc360 you can generate 20-30 TB of (native) data easily in just one year. And you will need fast access to this. This is huge especially for smaller offices.

https://openpointcloudformats.org/ : )