Hello,
I am a new user of CloudCompare and I am doing a study on a point cloud of a room interior created form one laser scanner. I've recently noticed that by playing with the scalar field I was able to see some different materials present in my point cloud, which it means that for example some curtains in my room that composed of wool and silk in different areas came with different colours under the scalar field although they are part of the same object.
My question is what exactly is the scalar field? what can I measure with it? and Where can I get more information on how to play with it?
I want to understand if I can trust on this split between two materials composing the same object under the same surface, I found this is very important for my work but I really need to understand what is the scalar field and how it works.
Thanks,
Saclar Field?
Re: Saclar Field?
Hi,
scalar fields are a very usefull and powerfull part of CC - they are simply attributes that where generated by your scanner or the scanner software. You can also generate new scalar field or calculate scalar fields in CC. A good example is that you can convert the x,y, z coordinates into scalar fields and e.g. color your point cloud using the z-coordinate. Your scanned room point cloud will then be gradually colored e.g. from blue for lowest points to red for highest points. The phenomenon that you can distinguish different materials might be because the scalar field that you currently use to color your point cloud is using the reflectance/intensity value that your scanner recorded. The laser beam that bounces back from the scanned object and that your scanner will record has a certain intensity which is lower than when it originally left the scanner. There is a number of factors that have an influence on this intensity value including the distance to the object (which is obviously used to calculate the distance. But the intensity is also depending on the material of the objects surface. In this respect moisture is very important - moisture will absorb the laser beam. Therefore, e.g. a wall build of granite will have a higher intensity than a wall build of moist clay etc.
The trouble with this intensity value is that it is easy to compare neighboring points in a point cloud, but because of the different distances to the object throughout your point cloud you need to do a distance correction to obtain homogeneous intensity values for the same material throughout your point cloud. You will notice that generally objects that are closer to the scanner will appear with higher intensity. To make this a bit more complex, most scanners have various filters build in that reduce or enhance the intensity to be able to have a good signal for a large range of distances. This means that the intensity vs distance is not a linear curve. Also, if you can effort to send your scanner to get checked by the manufacturer regulary, you will find that the filters will not stay the same and your distance correction need to be redone regularly. I heard that some scanners or scanner software packages allow for such corrections, but the cheaper ones don't (it is a black box). I hope this helps a bit.
cheers
mat
scalar fields are a very usefull and powerfull part of CC - they are simply attributes that where generated by your scanner or the scanner software. You can also generate new scalar field or calculate scalar fields in CC. A good example is that you can convert the x,y, z coordinates into scalar fields and e.g. color your point cloud using the z-coordinate. Your scanned room point cloud will then be gradually colored e.g. from blue for lowest points to red for highest points. The phenomenon that you can distinguish different materials might be because the scalar field that you currently use to color your point cloud is using the reflectance/intensity value that your scanner recorded. The laser beam that bounces back from the scanned object and that your scanner will record has a certain intensity which is lower than when it originally left the scanner. There is a number of factors that have an influence on this intensity value including the distance to the object (which is obviously used to calculate the distance. But the intensity is also depending on the material of the objects surface. In this respect moisture is very important - moisture will absorb the laser beam. Therefore, e.g. a wall build of granite will have a higher intensity than a wall build of moist clay etc.
The trouble with this intensity value is that it is easy to compare neighboring points in a point cloud, but because of the different distances to the object throughout your point cloud you need to do a distance correction to obtain homogeneous intensity values for the same material throughout your point cloud. You will notice that generally objects that are closer to the scanner will appear with higher intensity. To make this a bit more complex, most scanners have various filters build in that reduce or enhance the intensity to be able to have a good signal for a large range of distances. This means that the intensity vs distance is not a linear curve. Also, if you can effort to send your scanner to get checked by the manufacturer regulary, you will find that the filters will not stay the same and your distance correction need to be redone regularly. I heard that some scanners or scanner software packages allow for such corrections, but the cheaper ones don't (it is a black box). I hope this helps a bit.
cheers
mat
Re: Saclar Field?
Hi Mat,matknaak wrote: The trouble with this intensity value is that it is easy to compare neighboring points in a point cloud, but because of the different distances to the object throughout your point cloud you need to do a distance correction to obtain homogeneous intensity values for the same material throughout your point cloud.
Thank you very much for you awnser. I am really interested in explore this intensity scalar field to count the percentage of materials present in one object. And I want to do this in a scientific way.
Do you know if I can do the distance correction using cloud compare? If so what tools should I use?
Cheers,
Rui
Re: Saclar Field?
Hi Rui,
as mentioned before, if you are the lucky owner of an expensive scanner, the software that comes with the scanner may allready have options to correct the intensity values by distance, or atleast your scanner may allow settings without the use of such filters as discribed before. If you have a less advanced scanner with little support by the manufacturer (lile the case for me), you may have to make test scans to find out for a given material how the intensity changes with distance to the object. You simply chose some object and scan it every 5 m from 1 to 50 m distance and plot intensity versus distance. You will then learn that this is not a simple curve. If you are a math genius, you might be able to discribe the entire curve for these distances, but most likely there will be an intervall between 10 and 40 m where the curve (the relation between intensity and distance) is linear. So if you have figured out how to discribe the curve for relevant intervalls that you commonly use in your scans, you can use cloudcompares scalar fields arithmetics tool to calculate the corrected intensity values. Note that this is best done on point clouds of individual scans. In point clouds of several scans, neighboring points may come from different scanner positions and, thus, have very different distances to the object - this may be messy to clean afterwards.
For a start there is also some background in:
S. Kaasalainen, A. Jaakkola, M. Kaasalainen, A. Krooks, and A. Kukko,
“Analysis of incidence angle and distance effects on terrestrial laser
scanner intensity: Search for correction methods,” Remote Sens., vol. 3,
no. 10, pp. 2207–2221, 2011.
Höfle, B., "Radiometric Correction of terrestrial LiDAR point cloud data for individual maize plant detection", IEEE Geoscience and Remote Sensing Letters, vol. 11, no. 1, January 2014
Cheers
mat
as mentioned before, if you are the lucky owner of an expensive scanner, the software that comes with the scanner may allready have options to correct the intensity values by distance, or atleast your scanner may allow settings without the use of such filters as discribed before. If you have a less advanced scanner with little support by the manufacturer (lile the case for me), you may have to make test scans to find out for a given material how the intensity changes with distance to the object. You simply chose some object and scan it every 5 m from 1 to 50 m distance and plot intensity versus distance. You will then learn that this is not a simple curve. If you are a math genius, you might be able to discribe the entire curve for these distances, but most likely there will be an intervall between 10 and 40 m where the curve (the relation between intensity and distance) is linear. So if you have figured out how to discribe the curve for relevant intervalls that you commonly use in your scans, you can use cloudcompares scalar fields arithmetics tool to calculate the corrected intensity values. Note that this is best done on point clouds of individual scans. In point clouds of several scans, neighboring points may come from different scanner positions and, thus, have very different distances to the object - this may be messy to clean afterwards.
For a start there is also some background in:
S. Kaasalainen, A. Jaakkola, M. Kaasalainen, A. Krooks, and A. Kukko,
“Analysis of incidence angle and distance effects on terrestrial laser
scanner intensity: Search for correction methods,” Remote Sens., vol. 3,
no. 10, pp. 2207–2221, 2011.
Höfle, B., "Radiometric Correction of terrestrial LiDAR point cloud data for individual maize plant detection", IEEE Geoscience and Remote Sensing Letters, vol. 11, no. 1, January 2014
Cheers
mat