How to: Work with point cloud data in Civil 3D and InfraWorks

00:00

[MUSIC PLAYING]

00:06

RUSS NICLOY: Welcome, everyone to this webinar

00:08

of how to work with point cloud data in Civil 3D

00:12

and InfraWorks.

00:13

My name is Russ Nicloy.

00:14

I'm a Civil Solutions Specialist with MACER Technologies.

00:18

Before we start, I do want to share this safe harbor

00:20

statement.

00:21

Some things that are said during the session or in answers

00:24

to questions may end up being forward

00:26

looking at versions of software that are not yet released.

00:30

We want to make sure that you know not to make purchasing

00:32

decisions based on statements of possible future functionality.

00:38

As I said, my name is Russ Nicloy,

00:40

a Civil Solutions Specialist for MACER Technologies.

00:43

I've been in the industry for 26 years

00:45

with some time in GIS, survey, site, and utility design

00:50

and over a decade with the reseller partner of Autodesk.

00:53

I now run my own design solutions company

00:55

aiding organizations in the civil industry.

00:60

Just a word about the accelerators.

01:02

Autodesk accelerators are designed

01:04

to help your team stay ahead of the curve with the latest

01:07

workflows.

01:08

This is done with on demand courses, pre-recorded coaching,

01:12

and live coaching like what we're doing today.

01:15

To see a full list of the topics go to the Customer Success Hub

01:18

link.

01:21

Now, let's take a look at the learning objectives

01:23

and what you're going to learn here.

01:25

The statement is yesterday we did not

01:28

know why it's important to know about the characteristics

01:30

of point clouds and how this will affect our modeling

01:33

projects.

01:33

Today, we understand why this is important.

01:36

We'll be looking at typical point cloud Workflows

01:39

for Autodesk solutions, how point cloud data is used,

01:43

how point cloud data is collected,

01:45

why you need to the point cloud specifications,

01:49

point clouds from GIS sources, point

01:51

clouds from photogrammetry, and a summary of point cloud

01:55

considerations for using point clouds in your project.

02:02

This particular accelerator is going

02:04

to focus on several different parts of the project lifecycle,

02:10

first of all the design portion, then the field execution,

02:14

and also the handover.

02:19

Autodesk has created a powerful and easy

02:21

to use workflow for reality capture

02:23

in order to bring scan to BIM workflow to our customers.

02:27

Autodesk has partnerships with every major hardware vendor,

02:30

including both UAV or drone manufacturers and laser scan

02:34

manufacturers.

02:35

ReCap Pro has been designed for use by someone

02:39

with no previous experience.

02:40

So it's very user friendly.

02:42

And one click automated workflows

02:44

do tasks like scan registration, which stitches multiple scans

02:48

together into one point cloud and intelligent cleanup

02:51

services, which automatically remove unwanted noise,

02:54

like a person walking through an area that's being scanned.

02:58

And perhaps most importantly ReCap

03:00

integrates all of Autodesk key AEC design creation tools,

03:05

so you can bring that reality data into the data creation

03:08

tool of your choice, including Civil 3D, InfraWorks,

03:11

and Revit.

03:15

Here is the scan to BIM workflow process.

03:18

It starts with ReCap Pro to process the captured data,

03:22

moves over to InfraWorks to generate the surface

03:24

topology from the point cloud.

03:27

Then still in InfraWorks, it uses feature extraction tools

03:31

to extract line work data.

03:34

Finally, the data is then moved into Civil 3D,

03:36

which consumes that data and creates the design.

03:41

How is point cloud data used?

03:44

If you're only using point cloud,

03:47

you can use it as a visualization tool,

03:49

just to see what the area that you're working in.

03:52

You can use as a measurement tool,

03:53

measuring between actual points in the point cloud.

03:56

Or you can use it as a design background,

03:58

again, kind of that visualization

03:60

as you're working in an area.

04:03

Or you could use it as a basis for feature extraction.

04:06

This would be things like geometry or the surface

04:08

topography.

04:10

And obviously, you could do a hybrid.

04:11

You could do both, just use all those tools

04:15

that are listed there.

04:18

Some typical tasks in a point cloud data workflow

04:22

is you might be taking measurements from an area.

04:25

You could create points of features,

04:27

create 2D or 3D line work.

04:30

You can create a

04:33

create topographic surfaces, use as a design background.

04:37

You can calculate materials and volumes.

04:40

And you can monitor earthwork, construction, and commissioning

04:43

of a project.

04:46

Just a little bit about the different tools

04:48

at our disposal for collecting point cloud data.

04:51

LiDAR, which is actually light detection and ranging,

04:53

is a process of shooting out many beams of light

04:57

into an area and measuring their return.

05:00

Photogrammetry is the extraction of 3D measurements from images.

05:05

These are images that are overlapping and have

05:08

common points, so that measurements

05:10

can be taken from different vantage points.

05:14

And GIS repositories, now this is not

05:17

a form of creation of data, but it's

05:19

a valuable source of huge amounts of point cloud data.

05:25

This video is a high level demo lasting less than two minutes

05:29

covering the LiDAR to design portion of the scan to BIM

05:32

process.

05:33

NARRATOR: Utilizing LiDAR to design for the scan to BIM

05:36

workflow will allow you to capture

05:38

existing conditions in hours, instead of traditional survey

05:40

methods taking days in the same project.

05:44

Inside ReCap Pro, you can import mobile LiDAR scan data made up

05:47

of multiple scans into one model and start

05:50

to interrogate the data, then export out only what you

05:53

need to cover for your project.

05:56

Moving into InfraWorks, you can generate an existing condition

05:59

surface that accurately depicts the terrain

06:01

model of the project.

06:02

Just by changing from conceptual view to engineering view,

06:05

you're able to see the contour lines

06:07

on the surface that are generated from the LiDAR point

06:10

cloud.

06:10

This allows you to visualize and understand how the terrain lays

06:13

and what existing features and structures exist.

06:16

This existing ground surface is also

06:18

needed for linear feature extraction.

06:21

In InfraWorks, you have the ability

06:22

to extract linear features from the Point Cloud assembled

06:25

in ReCap Pro, for example, the center line of the road

06:28

striping, edge of pavement striping, et cetera.

06:33

You can quickly visually inspect the model

06:35

to be sure you have the necessary line work extracted.

06:39

This line work will be used for design inside of Civil 3D.

06:43

You can also interrogate the model data

06:45

with the cross-section view.

06:48

In this cross-section view, you can add additional brake lines

06:50

to points in the section view and automatically generate

06:53

additional linear features.

06:56

Lastly, here in InfraWorks, we can generate transverse lines.

06:60

These lines sample the cloud data at an interval in spacing,

07:03

so that you can export out a more lightweight but highly

07:06

accurate version of the cloud data.

07:09

Moving into design, here in Civil 3D,

07:11

we can take advantage of the highly accurate line work

07:14

and surface information that was extracted in InfraWorks

07:17

and apply it into our design.

07:26

RUSS NICLOY: So let's talk about point clouds

07:28

collected from LiDAR.

07:32

LiDAR, which is light detection and ranging,

07:35

is an active form of remote sensing using

07:37

beams of near-infrared light.

07:40

Topographic LiDAR pick up uses near-infrared light

07:44

and bounces that light work back.

07:46

Bathymetric uses water penetrating

07:48

green light specialized for that purpose.

07:51

Reflected signals can be classified.

07:53

So you can get information about the data that's returning.

07:59

LiDAR systems operate from terrestrial or land

08:02

based sources, airborne, or even spaceborne.

08:07

Different ways that you can collect LiDAR data

08:09

include the space borne from satellites, aerial,

08:13

which generally is aircraft at a certain elevation.

08:18

You can do it terrestrially from a static location

08:22

or from mobile units that are vehicle based,

08:26

backpack based as you walk through an area

08:29

or over a project or handheld for more localized

08:33

scanned information.

08:38

The new technology trend is the use

08:41

of ultralight laser scanners.

08:43

These are UAV or drone mounted scanners

08:46

that are integrated with the GNSS or geodetic control.

08:53

There is a link here for more reference information regarding

08:56

the ultralight LiDAR scanners for UAVs.

09:01

Point cloud specifications for your point cloud data

09:04

include density or point spacing, geodetic control,

09:08

and classification.

09:11

How you're going to use the point cloud determines

09:13

the point cloud minimal requirements,

09:15

also cost considerations, and can you use public data.

09:21

Point density includes how often those points

09:25

are shot in a certain amount of distance or area.

09:29

Density is critical for an automated feature extraction.

09:32

And resolution is a function corner point density.

09:36

You are going to want to know how you're

09:37

going to use your scan data to know how detailed you're

09:41

going to need to be.

09:42

Here's a chart or an explanation of the different types

09:45

of densities.

09:47

If you're doing just a basic surface,

09:49

maybe you're doing forest inventory,

09:51

and the surface isn't as important.

09:53

You can use the first option there,

09:55

which is a half point or a point every square meter.

09:59

If you're doing flood modeling or dam and water inundation

10:02

calculations, maybe you want to up it

10:04

to 1 to 2 points per meter.

10:07

Multipurpose data sets are generally

10:09

going to be in the 2 to 5 points per square meter.

10:13

Basic 3D models, you will want to use something higher

10:16

like a 5 to 10 points.

10:18

And a detailed 3D city model, you're

10:21

going to want closer to a 10 or more points per meter squared.

10:27

This is an example of the same surface

10:29

with three different levels of processing resolution.

10:32

When using a 30 meter resolution,

10:34

there are fewer points, and details are lost.

10:36

But the data is lighter weight.

10:39

At two meter resolution, we're seeing much more detail,

10:42

but the data is heavier for the process and storage.

10:46

We see something similar here, where

10:48

the area is important to the project have

10:50

more detailed design contours.

10:52

But the areas away from the design areas can be coarser.

10:58

You will want to know what your geodetic control is.

11:02

Is it low versus high accuracy point cloud, using horizontal

11:06

or elevation information.

11:08

You need to the units that are involved

11:10

and the geolocation or the coordinate zone of that data

11:14

and the methods for precise geolocation.

11:17

These include ground based control points,

11:19

active onboard INS, which is inertial navigation

11:23

systems that work with GNS or global navigation satellite

11:27

system.

11:28

Or do you have a simultaneous localization and mapping

11:32

system or a slam system.

11:36

Classification places scan points

11:38

into groups, such as points from ground surface returns

11:42

or vegetation or building and trees.

11:44

The different classes are defined using numeric integer

11:47

codes in the LAS file.

11:50

There is no standard for coding.

11:52

You must obtain the class codes from the data provider.

11:56

Not all LiDAR data is classified.

11:59

Extra sensors are required to do this.

12:01

Typically only airborne data is classified.

12:05

And classification is valuable for topographic surface

12:08

and feature extraction.

12:12

Not all point clouds are created equally.

12:14

Some have additional attributes.

12:17

Now some of these are also not recognized in ReCap Pro

12:20

or in InfraWorks.

12:23

They can be viewed and managed in Civil 3D, however.

12:27

There is a specialized workflow to extract ground surfaces

12:30

from point cloud data.

12:35

During the import process or attaching

12:37

of a point cloud to a file, you will have access

12:40

to a point cloud information.

12:43

This will include things like the point density

12:46

and any classification information that

12:48

might be involved in the point cloud data.

12:52

The NOAA data access viewer can display available LiDAR data

12:55

sets among other data for localized areas.

12:58

It is a great resource for broad uses of LiDAR data.

13:06

This data is aggregated from different sources

13:08

and may be in differing coordinate zones

13:10

or different levels of quality.

13:12

When accessing this data, it is important to understand

13:15

the specifications before you use the data.

13:21

Now let's look at point clouds from GIS sources.

13:27

Large area train mapping for many applications,

13:30

such as forestry, flood plains, hydrology, urban planning,

13:34

ecology, and volumetric calculations

13:36

requires surface and classification data.

13:39

These applications and more can take

13:41

advantage of combining LiDAR and GIS data.

13:45

Often there may be an existing source

13:46

for this information in GIS repositories,

13:49

both public and private, free or paid for.

13:55

Here's an example of point cloud data from a GIS data

13:57

source in the United Kingdom.

14:05

Another example is the ArcGIS Living Atlas source.

14:09

This data received improved resolution in 2019.

14:13

This data is available in InfraWorks and Civil 3D

14:16

through the Autodesk connector for ArcGIS.

14:22

Now let's look at point clouds created by photogrammetry.

14:27

What is photogrammetry?

14:28

It is triangulation using three dimensional measurements

14:32

from two dimensional images.

14:34

Now there need to be a lot of images through an area,

14:36

and they do need overlap so that you

14:39

can see different aspects of the area that you're measuring.

14:42

Survey control points are needed for geolocation, as well as

14:46

position and scale and accuracy.

14:49

The types of photogrammetry available

14:51

are terrestrial and aerial.

14:55

The aerial versions of these tools

14:57

can be things like aerial mounted cameras on aircraft

14:60

or on drones and UAVs or terrestrial

15:03

like the pole mounted camera in that picture there.

15:06

This high pole just provides a vantage point for the camera,

15:12

or you could use it to put it down

15:15

into subsurface structures, such as sewers and utility pits.

15:20

Note in that photo there's an orange cone.

15:22

There are actually two others around that stockpile.

15:25

These cones have a known exact location

15:28

and are used as GCPs or ground control

15:30

points, which will give scale and location to that pile

15:34

accuracy.

15:36

If you're capturing photos, depending

15:40

on the object or area that you're picking up,

15:43

you may use different methods.

15:44

If you're picking up vertical features, such as a building

15:47

or tower, you'll use a helical path orbiting

15:52

around the objects.

15:54

The ground features use a mow the lawn path, a back and forth

15:57

path, through an area.

15:59

The overlap of photos should be about 40% to give good results.

16:04

And you will need GCPs that clearly mark the targets

16:08

and appear in multiple photos.

16:11

The minimum number of GCPs is 4 or greater.

16:16

Some cameras have geotagging capabilities.

16:20

And that can geolocate the project.

16:22

But the results are very low accuracy.

16:26

It will provide a point cloud, but it won't necessarily

16:28

line up exactly.

16:31

This will distort elevation values.

16:33

For higher accuracy, you will want to use GCPs.

16:37

Now GCPs are required if you're going to geolocate the project,

16:41

provide scale, fix a planer position,

16:44

or add future photos to the point cloud.

16:48

Remember the data is only as good as the GCPs that you have.

16:54

This is a video of a high level demo lasting about two minutes,

16:58

covering the drone photos to design workflow for scan

17:01

to BIM.

17:03

NARRATOR: Utilizing drone to design

17:04

will allow you to capture existing conditions in hours,

17:07

instead of the traditional survey methods taking days

17:10

on the same project.

17:12

Inside recap photo, you can quickly

17:14

select the photogrammetry that was captured,

17:16

then add what are called GCP or ground control points that

17:20

will help to improve the accuracy of the model

17:22

and helps to ensure that you are tied

17:24

to a specific coordinate system.

17:26

The data that you received back from the cloud

17:28

is rich in content and accurately represents

17:30

what is there today.

17:32

Now we'll move into InfraWorks where

17:34

you can take advantage of the georeferenced image that

17:36

is produced from ReCap photo.

17:38

This image is high resolution and high in accuracy.

17:41

I'll call this the what is there today image.

17:44

You would not be able to get an image of this quality

17:46

without waiting for it for weeks.

17:48

Because we're using a drone, we had it in hours, instead

17:51

of weeks or even days.

17:53

In InfraWorks you can generate an existing condition surface

17:56

that accurately depicts the terrain model of the project.

18:01

Just by changing to the engineering view,

18:03

you're able to see the contour lines of the surface that

18:05

was generated from the recap photo point cloud.

18:09

This allows you to visualize and understand

18:10

how the terrain lays, so that you

18:12

can start to plan for your proposed project.

18:17

From inside InfraWorks, you can now access Autodesk Docs

18:20

and share your existing conditions model

18:22

with all stakeholders.

18:27

With the power of the Autodesk viewer inside of a web browser,

18:30

you can visualize, make markups, and communicate with all

18:33

involved in the project.

18:37

From InfraWorks, we're able to utilize the surface

18:39

data inside Civil 3D.

18:42

You not only have the surface to use

18:43

to start the preliminary design, but you can also

18:46

bring in the what is there today image that we

18:48

talked about earlier.

18:52

Once the data has been imported into Civil 3D,

18:54

you're able to start using design functionality to plan

18:57

out for instances like a new roadway location, utility

18:60

location, et cetera.

19:03

Take advantage of the dynamic functionality of Civil 3D

19:06

to start the design of your roadway.

19:12

RUSS NICLOY: So which system is better, photo versus LiDAR.

19:15

Well it depends on your needs.

19:17

In this case, we're looking at the same area in both an air

19:21

photo and a LiDAR image.

19:23

You'll notice in the LiDAR image below that we

19:25

can see a landslide area that has been obscured

19:29

by trees in the photo area.

19:35

As you're receiving point cloud data to use in your project,

19:38

you do need to ask these minimal questions.

19:41

What is the intent and purpose of the scan data.

19:46

What collection type was used.

19:49

Is the orthophotography required for the project.

19:53

Establish coordinate systems and units.

19:56

What survey control network and height accuracy is being used.

19:60

Is extracted data expected.

20:01

For example, do you expect a surface topography

20:04

with the point cloud.

20:06

What are the delivery file formats,

20:09

and how are the files going to be delivered.

20:14

Depending on your projects, there

20:15

are important questions that should be known

20:17

about your point cloud data.

20:18

What will the data be used for.

20:20

How will the data be used.

20:22

What level of data processing is appropriate.

20:25

Are classified data layers important.

20:30

The cost of the LiDAR project depends on the project scope,

20:33

location, boundary shape, point density, deliverables,

20:37

time frame, weather patterns and so on.

20:40

Remember all point cloud collection systems are not

20:43

the same, each of them is optimized

20:45

to provide data sets that can differ greatly.

20:48

More is not always better.

20:49

High density point clouds are not required for every project.

20:53

And huge file sizes and extreme computing power requirements

20:56

must be considered.

21:06

If you want to learn more about how LiDAR works,

21:09

here's a link to a slide deck of information about that.

21:16

Here are some reference links, including LiDAR 101 from NOAA,

21:22

a reference on accuracy between LiDAR and photogrammetry,

21:27

references on LiDAR versus photogrammetry,

21:31

and a reference for new ultralight LiDAR

21:34

scanners for UAVs.

21:40

Let's look at some additional information on laser scan uses.

21:45

The possible uses are endless and really

21:49

just depends on the project that you're working on.

21:52

It's great for architecture, essentially any situation where

21:55

numerous measurements are necessary, especially

21:58

in hard to reach areas, also in construction projects,

22:11

or in survey, or for making models

22:18

for media and entertainment purposes,

22:26

and for forensics, police and accident investigations.

22:35

This is part two of how to work with point cloud data

22:37

in Civil 3D and InfraWorks webinar.

22:40

In this session, we will be looking

22:41

at using ReCap and ReCap Photo in a scam to BIM workflow.

22:47

So the learning objectives for this portion are going to be,

22:51

first of all the statement of yesterday

22:53

we did not know how to make point cloud data ready for use

22:56

in Autodesk products and other review and design solutions,

22:60

today we can do this.

23:02

We'll be looking at typical point cloud workflows

23:05

using Autodesk recap file import and export types, project

23:10

set up, indexing and registration,

23:12

adding control points, view, review,

23:15

and measure, crop and cleanup, and exporting regions.

23:22

In the scan to BIM concept workflow this session,

23:25

we'll be looking at the compute phase.

23:27

We'll look at automatic scan registration

23:30

and intelligent cleanup of data.

23:35

Now a little bit about ReCap and ReCap Photo.

23:39

These are Autodesk applications that

23:41

consume point cloud data from reality capture or RCS files.

23:46

ReCap and ReCap Pro are related tools.

23:52

ReCap Photo is separate from that,

23:54

and we'll be working with images to capture that same data.

24:03

Here's a look at the ReCap solution workflow.

24:07

First of all, there is the import point cloud data

24:10

and set the advanced settings portion of this.

24:13

Then you'll move on to index and register to aggregate

24:17

the data of the scans.

24:20

Then you'll perform visualization

24:22

and measurement and cleanup.

24:26

And finally we'll go in and create an export regions.

24:35

Here's a two minute video on the project setup and import.

24:39

When you first start your project in ReCap,

24:42

you'll be importing scan files into the project

24:45

but also creating the settings file

24:49

or the settings information inside the project.

24:51

So we're going to come in here and start a new project.

24:54

It will immediately ask us, do we want to import point cloud

24:57

information or photos to 3D.

24:59

I've got point cloud information I'm going to bring in.

25:03

And this is the redtail project.

25:09

And it's also talking about the folder

25:10

where it's going to be dropped.

25:19

And I need to go over to the Civil 3D projects

25:22

to my redtail project.

25:27

And I'll select that folder and then proceed.

25:31

Now from here, you can browse around for files

25:34

or add files to import.

25:37

But I'm going to use the drag function.

25:38

I just have a Windows Explorer folder with all my LAS

25:41

files here.

25:42

I'll shift select those and drag those into this folder

25:45

or into that area there.

25:47

Then while it's doing that, it's asking me what kind

25:50

of scan settings we want.

25:53

I can clip the points down and say,

25:55

if the intensity isn't high enough,

25:56

maybe the points aren't real and kind of filter those out.

25:60

I can also say that, well I don't

26:01

want points that are too intense on too big of a return

26:05

and clean that out.

26:07

I can also go to the Advanced button

26:08

here and look at the decimation.

26:11

Basically, this is taking points out for distance,

26:15

so kind of cleans up the scan, when you have too much data.

26:19

It also looks at the coordinate system that's involved here.

26:23

I'm going to go ahead with those settings right there.

26:26

I'll click on Import files.

26:29

And then all of those LAS files are represented here,

26:33

and will begin to import.

26:37

It is important to transform your data

26:39

to the proper coordinate zone.

26:40

You may be transforming between a standard coordinate zone

26:43

and a custom coordinate zone.

26:45

ReCap uses the Autodesk geospatial system library

26:48

found in the C drive program data Autodesk

26:52

geospatial coordinate systems.

26:55

Custom coordinate systems are found at the local app data

26:59

folder location, Autodesk user geospatial coordinate systems.

27:07

XML files are used to share custom coordinate systems

27:10

with these commands-- the map CS library

27:14

export to export the file and the map CS library import

27:18

to import that XML.

27:23

You will notice as we move through the process

27:25

here that there are two different types of files

27:27

that we're working with.

27:28

RCS, which are reality capture scans

27:31

those are the actual scan files, the individual scan files.

27:35

RCP is the reality capture project file.

27:39

And this is a small file that directs the information

27:43

from the other RCS files.

27:46

Usually there are more than one RCS file involved.

27:49

And RCP kind of directs that.

27:53

Other exports from ReCap that are

27:54

available-- you can export the scan files to a single file,

27:58

like an E57 or PTS or PCG file.

28:02

But no, because all these points are contained in one file,

28:06

they can end up being very large.

28:11

Exports from ReCap photo are going

28:13

to be into things like FBX and TIF, GeoTIFF, ortho photo.

28:24

One of the first steps during this process

28:26

is indexing and registration.

28:31

Indexing just reads the raw point cloud data

28:33

and converts it into RCS files.

28:37

Registration merges or stitches multiple

28:39

scans together into one aggregated point cloud.

28:43

Now there are two ways that it does this, auto registration

28:46

or manual registration.

28:47

The auto registration becomes available

28:50

once all the files have started importing.

28:53

You simply click Register scans, the button

28:55

in the lower right corner.

28:57

And that starts that process for you.

28:59

Manual registration in some instances, auto registration

29:03

will not be able to find enough matching data.

29:06

And so in these cases, you will be

29:09

prompted to manually register the unregistered scans.

29:12

Manual registration can be completed through the cloud

29:15

to cloud technique, whenever auto registration fails or even

29:19

in place of it.

29:22

Manual registration is available at any time

29:24

during the registration process.

29:26

Electing to leave the auto registration

29:28

compiles any registration already

29:30

completed and allows you to finish the registration

29:33

process through the standard manual workflow.

29:36

The tool can be accessed during the registration process

29:39

from the bottom toolbar.

29:40

Manual registration is also only available in ReCap Pro.

29:47

There are several tasks that you have in ReCap.

30:04

There are several tasks that you have in ReCap.

30:12

You can visualize the point cloud

30:13

using colorization of points due to different factors,

30:17

like intensity or elevation or even by point classification.

30:23

You can also measure between points in the cloud.

30:30

You can clip points from certain views

30:32

or crop them from the data.

30:34

And this will help reducing the size of the point cloud

30:37

that you're working with.

31:07

Then you can export those points in the area of interest

31:11

for use in other platforms.

31:24

While ReCap Pro handles these larger point clouds well,

31:28

you may still want to create regions to access

31:30

or focus on more localized areas of the scan.

31:37

This is useful for exporting smaller areas out

31:40

of ReCap Pro for other applications.

31:43

You can export these regions out as their own RCS

31:46

file or their own RCP files.

31:51

This is a look at the ReCap photo solution workflow.

31:55

First of all, we're going to upload photos.

31:58

Then we'll move over and add ground control points or GCPs.

32:04

Then you'll receive the processed RCS files and ortho

32:07

photo from that combination.

32:11

Then you'll import point cloud regions and ortho photos

32:14

for use in other design applications.

32:19

You can create the point cloud and ortho photos

32:21

from a series of photos that you've

32:23

imported into ReCap photo.

32:26

This is the front page of recap photo

32:27

that manages the different projects

32:29

that you're working in.

32:32

In this short video, we will see the actual collection

32:35

of photos in a difficult area of a utility pit.

32:57

Good lighting and a series of photos

32:59

is needed for the later processing.

33:07

In ReCap photo, these images are indexed and stitched together

33:10

to create a model of the utility pit.

33:23

That model can be accessed and measured as needed.

33:26

So you won't need to return to the site.

33:40

If you want more about ReCap, follow the link

33:43

to the self-paced learning at the Autodesk knowledge support

33:46

page for ReCap.

33:52

Here you will find technical support articles, Autodesk

33:55

University sessions, YouTube videos,

33:57

and many other information sources

33:59

to further learn about these tools.

34:05

This is part three of how to work with point cloud data

34:08

in Civil 3D and InfraWorks.

34:12

In this session, we'll be looking at point cloud data

34:14

in Civil 3D and InfraWorks.

34:18

The learning objectives for this section,

34:20

and again, the statement is yesterday

34:23

we did not know how to officially use

34:24

point clouds for modeling in Civil 3D or InfraWorks.

34:27

Today we know how.

34:29

We'll be covering: attaching and management

34:31

of point clouds, elevation, intensity, and classification,

34:35

point cloud tasks, including measuring, cropping, sections,

34:38

and slices, extracting simple geometry,

34:42

extracting vertical and linear features, accurate

34:45

surface creation and point clouds from GIS sources.

34:52

The first step to creating a surface from a point cloud

34:54

is to know where the project location is.

34:57

Assigning a Cordon system to the project,

34:59

make sure that the scale and rotation

35:01

are correct alongside that location.

35:05

Since the point cloud will be a large number of points,

35:07

the surface we are creating may not need all of those points

35:11

to be correct.

35:12

If your surface does not use every point

35:14

but uses enough to be accurate, you

35:16

will see a huge efficiency gain as you work with the project

35:19

later.

35:23

A little bit about classification,

35:24

not all point clouds are created equally.

35:28

Classification is an additional attribute.

35:31

It does require special data collection

35:33

equipment and specifications.

35:37

It may not be recognized by ReCap Pro either.

35:40

You can view and manage this data in Civil 3D.

35:44

It does require a specialized workflow

35:46

to extract individual classifications.

35:52

If you're going to make a surface from ground

35:55

classification points, there are tools specifically made

35:58

for this.

36:00

Civil 3D can filter out all information

36:03

other than the desired point class.

36:06

For this, we're going to use the command map, create PC surface.

36:11

This will open up a dialog box that

36:14

will help us create the surface from the point cloud.

36:17

It's going to create it as a TIF file.

36:19

What you'll do is you'll click on the Filter

36:21

in the lower left of the dialog box,

36:23

and then you'll select the ground option.

36:28

This will create a TIF file using only the ground points.

36:33

Later on, you'll be able to create a surface in Civil 3D

36:36

by importing this as a DEM file or digital elevation model

36:39

file.

36:43

There are filtering methods that you

36:45

can use for creating surfaces from point clouds.

36:50

You will want to understand how to apply

36:52

a filter before you start.

36:55

In most cases, the no filter option

36:58

may be the one that you need to go with.

37:01

A little bit about the filter options, the planar average

37:04

specifies which points should be filtered as non-ground points

37:07

by calculating an average between the elevations

37:10

of the points and then filtering out

37:12

the points that are at an elevation above this average.

37:15

In that illustration, the top illustration,

37:17

the blue points represent the points that are filtered out.

37:20

And the red points represent points

37:22

that will be included in the surface.

37:25

The Kriging interpolation option interpolates

37:29

new data points to build curves and then filters out

37:32

the points that are at an elevation above the curves.

37:34

In that second illustration, the blue points

37:37

represent the points that will be filtered out.

37:40

And the red points will represent

37:41

the points that are being included in the surface.

37:45

In the last image, there shows the no filter option,

37:48

which doesn't apply any filter.

37:50

And so all the points in the selected point cloud

37:53

will be used in the resulting surface.

37:58

This is a minute and a half video on ReCap

38:02

and how ReCap can be used to register an index huge point

38:06

cloud data as RCP and RCS files.

38:10

NARRATOR: First we set the coordinate system.

38:13

And then we went into the locations of the point

38:16

scans to analyze the data, specifically

38:21

the size of the point RCS files that were imported.

38:25

We can further analyze this data by going in and changing

38:29

the colorization.

38:30

Right now it's an rgb or true color.

38:32

But we can look at the elevations or the intensities

38:36

of the shots.

38:38

We can also change the background color,

38:40

if that helps us identify where there are holes in the surface

38:43

or just makes the point cloud easier to see.

38:51

There are some areas in here that we have set so

38:53

that we can zoom over to them.

38:55

So we can look at the top down plan view or specific locations

38:59

in the project area.

39:03

In this case, we're going to go in and make a dimension in here

39:07

between points, so that we can see how wide this roadway is.

39:13

This measurement can stay there.

39:18

Or I can go in and hide it or just delete it.

39:21

So it won't be available later on.

39:26

After analyzing the area here, I do

39:28

have some regions set up so that I can

39:31

identify where they're located.

39:33

You can see them highlight in the screen there.

39:35

I'm going to hide the two regions that I don't need

39:38

and then leaving just the one and exporting that out

39:41

to an RCS file.

39:44

You'll notice that during that process,

39:46

it says where it's going to save.

39:47

And it will unify this data.

39:50

Because it's a smaller data, that thinking

39:52

bar there is shorter than it would be if we

39:54

had to do the whole setting.

40:05

RUSS NICLOY: Now from there, we're

40:06

going to build the surface from classified point cloud data.

40:10

We're in Civil 3D, where we're setting the coordinate zone

40:15

for the project.

40:17

I'm setting the coordinate zone category,

40:20

which then gives us a list of the coordinate zones which

40:22

I'll select from.

40:24

If you know the coordinate zone code,

40:26

you can just type it here directly.

40:34

Then we're going to access a tool called the PGP file.

40:38

This is the aliases for commands that are involved.

40:45

There are several commands that we

40:47

have that are not in this list, and I'd

40:49

like to access them more quickly.

40:53

I'm going to type in LOD on or a level of detail on.

40:60

And then maintaining the typed versions of these commands,

41:04

I'll also add the LOD off but then

41:07

add those command typed versions in this list.

41:10

These are commands that I would be typing at the command line

41:13

otherwise.

41:14

I'll come down into this area and put

41:16

in PCM, which will open up my point cloud manager.

41:23

This step is not necessary, but it

41:25

can be helpful when you're typing commands in often.

41:37

I do have to re-init or re initialize Civil 3D

41:41

and make sure the PGP is turned on.

41:47

Now those commands will be ready when I need them.

41:52

Now we're going to go in and attach the point cloud

41:55

information.

41:56

I'm going to make sure that I'm on the wire frame setting,

41:59

because the 2D wire frame will not display the point cloud

42:03

data.

42:13

I'll go to point cloud and attach and then

42:18

bring in the RCP or actually the RCS

42:20

file that I exported from ReCap Pro.

42:30

In here, I'm going to show the details,

42:32

so that I can check to see the classifications are turned

42:35

on and available in this set.

42:38

I am going to set them to zero, because my coordinate systems

42:41

will line this up with my project.

42:51

And from there, I'm going to go in, because this is geolocated

42:55

and I have put in a coordinate system.

42:56

I can turn on the map aerials for this area

43:01

and see if this lines up.

43:02

Now there are some areas where it doesn't exactly line up.

43:06

But in the most case, most of it does line

43:09

up right where we need it to.

43:17

I'm opening up that point cloud manager

43:19

again using the typed command and identifying

43:23

where that project area is.

43:25

I can turn it on and off, if I don't need to see it

43:28

during certain processes here.

43:34

If I select that point cloud, I'm

43:37

grabbing the outside edge of it.

43:39

I'm going to go over to display and make sure

43:41

that the point sizes are smaller using that slider.

43:46

I'm also going to go into the scan colors

43:48

and go down to classification.

43:53

There I can see everything is colorized

43:54

per the classification.

43:56

I'm going to go into the color mapping

43:58

to make sure I know what those colors are.

44:00

I'll select all of them and then uncheck them and make sure

44:04

that they're all turned off.

44:05

Notice that, that basically turns everything off.

44:08

I am going to turn back on the number two,

44:10

the ground points, those points that

44:14

are classified as ground shots.

44:22

Now I'll select that frame again for the point cloud

44:26

and go to Create surface from point cloud.

44:29

This is the Civil 3D tool.

44:31

I'm going to name that surface.

44:35

And notice that I'm using 100% of the points

44:37

and how large that point set is.

44:41

I'll go into the filtering, and in this case,

44:45

I'm going to use no filter and then

44:47

create the surface from that.

44:50

This is going to publish in the background.

44:52

So it will return when it's done.

44:58

You can zoom to the point clouds,

44:60

if you don't know where they are.

45:10

And now we'll take a look at the contours

45:13

that were made from that surface from that point cloud.

45:21

You can see here that there are areas

45:23

where there are a lot of contours indicating

45:26

that there may be things higher than the actual surface.

45:31

But you can also see that there are

45:33

areas where the contours are laid out nicely.

45:36

There may be points that stick up in certain areas,

45:39

giving us that kind of island feel of these contours.

45:46

So let's now take a look at the map create PC surface tool.

45:55

First, I'll take a look at the properties and statistics

45:57

of the first surface to see the number of points.

46:06

Then I'll turn that style to no display.

46:14

Now we're going to use the map PC create surface tool,

46:20

select from the point cloud, identify the level of detail

46:23

that we want to use.

46:25

In this case, we'll use 100%, and then

46:28

identify the TIF file that we want to create from this.

46:32

We're going to call this project area GRND.

46:34

It is a TIF.

46:36

We'll save it, and, we will filter by the ground point

46:40

classification.

46:44

The layer that this is put on in Civil 3D

46:47

will be the project area-GRND.

46:51

That processing takes a minute, so we will return.

47:04

Now I'm going to go into the create surface from DEM tool,

47:08

find that TIF file that we just created and open that.

47:15

Now that creates a surface in our surface list.

47:21

And there you can see the contours

47:23

that are built from that.

47:25

These contours are much smoother than the surface

47:28

that we built earlier.

47:30

If I go into the statistics, you can see the number of points

47:34

is much lower than the original.

47:42

If we go into the point cloud manager,

47:44

we can turn the point clouds off,

47:45

so that we can see the surface.

47:48

I will also go in and change the viewing style to 2D wireframe.

47:53

Notice that it does give you a warning message

47:55

that the point cloud will not be visible in the 2D wireframe

47:60

visualization.

48:02

I'll also turn off the layer that the point cloud

48:04

object is on.

48:05

So even the frame is not visible.

48:08

But now we can see the surface, as it

48:10

will be while we work with it.

48:15

I'll go back over to the surface properties

48:18

and change the original surface back to 2 in 10 contours

48:22

to match the new TIF surface.

48:24

And there you can see the differences

48:26

between the surfaces.

48:33

I'm going to open up a second viewport,

48:37

so that we can compare and contrast and visually analyze

48:40

this.

48:41

I'll draw in a line, so that I can make a quick profile

48:47

in an area.

48:50

We'll make sure to set different styles, so that we

48:53

can tell the surfaces apart but then place that quick profile

48:57

in the other viewport.

48:59

There you can see the differences.

49:01

The original surface has more elevation differences,

49:05

while the TIF surface is smoother, more realistic

49:09

to the area.

49:11

There you can see that those contours did indicate

49:14

a very large rise in the surface, probably

49:16

some visual noise.

49:23

So now we're going to look at the two surfaces

49:27

to compare and contrast them.

49:33

I'm changing the style of the original surface

49:36

to the tin triangles.

49:38

That way I can see where those triangles locate.

49:41

I'll zoom in to take a closer look.

49:47

Now I'll go in and change the other surface

49:50

to the triangle style, so that we can analyze that in relation

49:55

to it.

49:56

Notice that these are not in a grid, that they

49:58

are a tin or Triangular Irregular Network.

50:04

And you can compare them to the original one, when

50:06

they're overlaid like this.

50:13

Now there are times when you'll be

50:14

working with non-classified point cloud data.

50:19

If you look at this, the larger image

50:22

there shows an overall project of a mobile scan

50:25

along a roadway.

50:26

And then we're zoomed in on an inset of a roundabout area.

50:31

The region of the roundabout, I'm going to call it roundabout

50:37

So you can see that there's an overwhelming amount of data

50:40

in this area here.

50:42

If you were to build a surface in InfraWorks versus Civil 3D,

50:46

you would see that this information is handled

50:49

differently by both products.

50:54

Notice here that we have an RCS file for our project area

50:59

and what that size is, it's quite large.

51:09

Now in InfraWorks, we've got the point cloud region in the area.

51:13

We're going to turn on some point cloud themes,

51:16

so we can analyze for elevation, as well

51:19

as what the classifications of those points look like

51:21

and where those are.

51:24

I'll return over to the point cloud

51:25

tools to create a point cloud terrain.

51:29

I'm going to use the project area there.

51:31

I'm using the optimum settings on all of the items there.

51:35

I am also going to override the model point cloud

51:39

and create my own.

51:40

I'll use all the points in this and then start the processing.

51:44

This will take a minute, and we'll return when that's done.

51:48

Now that terrain has been created underneath that point

51:50

cloud.

51:51

So I'll export the point cloud.

51:55

I'm only going to export the point clouds, not

51:57

the linear features.

52:01

Then I'll set the coordinate system

52:03

and make sure that I know where this file is

52:05

going to be located.

52:15

Now in Civil 3D, I use the attach function

52:19

to bring in both the original project

52:22

area and the new project area.

52:30

I'll turn off the point cloud that I don't want.

52:39

And then from there, use this other point cloud

52:41

to create a surface from that point cloud.

52:45

This is the from InfraWorks point cloud.

52:50

Going to bring in all the points.

52:52

In this case, I'm going to use the Kriging interpolation

52:55

filter and let that process in the background.

53:02

When that's back, you can zoom to where

53:04

that surface is created.

53:06

And you can see the contours there needs

53:09

some triangles removed from it.

53:12

I'll turn on the other point cloud

53:17

and then create a surface from this point cloud that was

53:19

brought into Civil 3D directly.

53:24

This time I'll use no filter and create that surface.

53:35

Now I'm going to check the statistics of the all points

53:39

surface and compare it to the properties of the InfraWorks

53:45

surface.

53:48

You can see the InfraWorks surface is much, much smaller.

53:56

Now I'll turn off the both point clouds and the point cloud

53:59

manager.

54:04

I'm also going to turn off the layers where

54:06

those point clouds are stored.

54:11

That way we're just looking at surface data.

54:18

I'll create a line, so I can make a quick profile

54:21

to study this area here.

54:40

I'll drop that profile in this area here.

54:45

Now as that line goes through that area of high contours,

54:50

I can see that that's a straight up point.

54:52

I think that's actually the massed arm from a light.

54:57

So I'll move that line around and take some looks

54:60

at other parts of this point cloud area

55:02

or the surfaces created from the point clouds.

55:08

In the InfraWorks process, you will be setting the processing

55:11

rules.

55:12

In here it's.

55:13

A good idea to use optimum.

55:14

This is a good balance between all points available

55:17

and the points necessary to produce an accurate terrain.

55:21

As a note, depending on the density of the point cloud,

55:24

there may be little to no difference in the generate data

55:27

option, if you use lightweight versus all points.

55:35

You can see here in one image, we

55:37

have made the region with only the ground surface points.

55:40

And in the other image it includes all the points

55:42

in that area.

55:44

The InfraWorks terrain.

55:45

Will understand that the more vertical points do not

55:47

belong in a surface and will not include those points.

55:52

Here you can see the comparison of an InfraWorks surface

55:55

versus the Civil 3D surface.

55:57

First, notice the size of the InfraWorks surface

55:60

has a much smaller number of points.

56:02

This is due to its own filtering process.

56:05

Then notice that there are irregularities included

56:08

in the Civil 3D surface, due to points from vertical objects

56:11

being included.

56:15

It's a real possibility that using a point cloud

56:18

will create a very large surface.

56:23

In AutoCAD Civil 3D you can build, edit, and save

56:26

large surfaces.

56:27

The term for large surface applies

56:29

to a surface with a number of points exceeding one million,

56:32

generally one million points for a grid surface

56:35

or a snapshot of a tin surface and two million points

56:38

for a tin surface.

56:40

AutoCAD Civil 3D has a mechanism for handling operations

56:44

with large surfaces.

56:45

When you save a drawing containing a large surface,

56:48

the data pertaining to the large surface

56:50

is written to a separate file that's

56:52

saved in the same location where the surface drawing resides.

56:56

The tin surface comparison files have an MMS file extension,

56:60

and the grid surface files have a GRS file extension.

57:04

The file name of a surface comparison file has the drawing

57:07

name underscore surface object handle .MMS format.

57:13

By the way, these files do have to be in the same location

57:16

as the surface file, the file that holds the surface.

57:20

So every time you move that drawing

57:22

file that has the surface, you will

57:24

need to move these companion files to that same location.

57:33

Here you can compare the file and object sizes.

57:35

In the top image, you can see how large the RCS point cloud

57:39

file is compared to the TIF or GRS files that result--

57:43

those surfaces that result from that.

57:46

You can also compare the number of points

57:48

from the Civil 3D surface to the GRS surface.

57:52

And finally, notice the surface and contour differences.

57:56

Despite much less data, the surfaces

57:58

are reasonably close in outcome.

58:05

We do suggest a temporary first step to your surface,

58:08

and that is to export it as a DEM file.

58:11

Why?

58:11

Well, for performance.

58:13

Use the map create PC surface command from the map 3D tools

58:17

to get that surface from the classified points.

58:23

You can then export this surface with a large grid spacing.

58:26

This creates a TIF file of the surface that can be

58:28

imported into your design file.

58:35

Boundaries will be useful in reducing points in the surface

58:38

from areas that should not be included.

58:40

You can use outer to remove points from outside the surface

58:43

area or hide to remove points from areas inside the surface

58:46

boundary.

58:51

So which is better for intent and purpose.

58:53

A comparison of the surfaces shows

58:55

that we are not losing accuracy of the surface,

58:57

but we will be gaining efficiency

58:59

for working with the surface later on.

59:02

Depending on your project needs, this

59:03

could make working in the area much easier.

59:11

In Civil 3D, you can analyze areas of the project

59:14

by looking at sliced views of the point cloud.

59:18

In this case, I'm going to make sure

59:19

from the point cloud manager, which point cloud is on.

59:24

I'll select the point cloud and choose the Create section tool.

59:29

Draw a line through, and that's where the section will be.

59:32

I'm actually going to create a second viewport,

59:34

so I can look at it in this view.

59:36

I'll select that section and say,

59:38

to change that view to the UCS of the section.

59:42

I do have to find where that dropped in.

59:44

But there you can see that section.

59:47

It's actually seeing some data behind the section.

59:50

So I'm going to go in and change the section to a slice.

59:54

This will give me a back clipping plane, as well as

59:56

the front clipping plane, so that I can shift that down

60:01

and not see so much behind it.

60:06

I'll turn on the imagery, so that

60:16

can see what that looks like in comparison to the imagery.

60:23

And now I'll move that around to find different areas

60:25

of my design area.

60:43

When you have a sliced view of an area,

60:45

you can use that to measure between point locations

60:48

or for further analysis.

60:52

This way you don't have to have the Civil 3D

60:54

object built to start analyzing distances in areas.

61:01

I'm going to create that section through this

61:04

and change that over to a slice and then shorten

61:08

the back clipping plane, so that I

61:10

can see just a very narrow area of our project, then zoom in.

61:15

And generally, I can see where the curb and gutter is.

61:21

I'm going to make sure that my object snap settings have

61:24

the 3D Object Snaps turned on.

61:29

And I'm snapping to nodes.

61:40

Then I'll go in and start the distance command,

61:42

measuring from one point to another point

61:45

in the general location of where those curb and gutters are.

61:50

You can see here that I've got a distance between those,

61:53

generally.

61:54

There is a little bit of z elevation difference there.

61:57

So I wasn't exactly on.

61:59

But I'm close enough to understand

62:01

what the size of that road is.

62:08

The tools that are important to that process

62:11

are that we set the UCS or user coordinate system

62:14

to the section plane and use the plan

62:16

function to move our view to that UCS. Also

62:21

you noticed that we turned on the 3D Object Snaps,

62:24

so that we could snap the points that we

62:26

need to measure between.

62:30

An example of this type of measurement function

62:32

is if you were referencing a PDF of an object,

62:36

like equipment or in this case a train car at a platform,

62:40

so that we can measure the point cloud

62:42

against those other objects.

62:49

In this example, we're going to measure the wall

62:52

height from a point cloud.

62:54

Now we're going to zoom over to where the project location is

62:57

and export that data out as an RCS file,

62:60

so that we can use it later.

63:05

Then you'll see that we're going to bring this into AutoCAD

63:09

and insert this data, attach that RCS file,

63:14

go over to the project and find it.

63:17

There's that region.

63:22

And I can see the extra information there.

63:24

I'll bring that data in.

63:28

And I'm going to use the view cube

63:29

to find the location that will best identify the wall that I

63:33

need to see.

63:37

Use the distance command and snap

63:39

to two points that represent the height of that wall.

63:45

This way, I don't need to go back

63:47

to visit the site to get information from that area.

63:58

In this example, the point cloud is not

64:01

aligned to the project area.

64:07

We're going to use the AutoCAD align function

64:09

to identify four locations in the point cloud

64:14

and where those locations belong in the drawing file.

64:32

This is going to align the point cloud

64:34

with the imagery and other project objects

64:39

in both size, scale, location and rotation.

64:46

Notice that I do have to go and change the 2D wireframe

64:49

to the wire frame or 3D wireframe

64:52

to see the point cloud in there.

64:54

I can compare it to the imagery and notice

64:58

that it is lined up right where we need it to be.

65:10

If you want more self-paced learning on these topics,

65:12

remember to visit the civilcommunity.a

65:14

utodesk.com/#learning.

65:25

Here's some reference information

65:26

regarding file and object sizes of tin versus TIF and point

65:31

cloud sizes.

65:36

If you want additional resources,

65:38

we would like to encourage you to visit the Customer Success

65:40

Hub for more resources like courses and learning paths,

65:44

recorded coaching sessions, and more live coaching topics

65:47

like this one.

65:48

Just follow the link to customersuccess.autodesk.com.

65:52

Thank you all for joining us today and have a great day.