3D laser scanners

I use the imported mesh in Rhino as input for creating NURBS surfaces. The surfaces are used for calculations. The complexity of the process for creating surfaces from the mesh depends on the shape and complexity of the original surface, the accuracy and level of detail desired, and the quality of the input mesh.

I don't think masking tape would work. The "texture" needs to be visual, not tactile. A friend has used a spray of powered pigment in water to create a "splatter" effect which works.

 

This is the process I'd like to know but I guess it is to complicate to be explained Here . Thanks DCockey

 

I was thinking different color masking tapes.

 

Unless you do a lot of scanning, it will be a lot easier and cheaper to rent a scanner. Usually you will also get better results hiring the work out. Learning the software and how to use the equipment takes some time. We use different types of 3D scanners for above and below water. For example, BlueView can scan the hull of a boat that is in the water. However, the scanning head alone is $170,000.

 

I think I'll try Metashape first. Digitizing stick-built boats for production is the majority of my job. Hiring out would also include firing myself.

 

I am sure you are aware, but just in case GeoCloud https://www.geocloud.work/app/#/add they have Metashape Pro on supercomputers that can be rented by he hour.

 

I developed a system to do laser measurement of ceramic parts because the ceramic parts have significant tolerances that one has to compensate for in the assembly process. In my application it was all about speed and accuracy. When the part to part cycle time is 30 seconds and measurement error has to be less than .004", then there are some very good (if expensive) laser triangulation sensors made by Keyence in Japan. They have a 21 bit parallel data bus which can output data up to 50khz so one does not have the problems associated with serial communication. We used the encoder pulses on the motion control system to synch the data acquisition, that way there is never any doubt regarding to which position the dimension belongs.

There are many pitfalls associated with such systems that have to be avoided if the data is to be meaningful, one of which is the linearity of the sensor, which for longer focal lengths and longer measurement ranges tends to be considerable... The problem of latency in the measurement process, relative to the travel velocity of the sensor is another that I have touched on.

For our application, we evaluated systems that were marketed for up to $100k and found them to be lacking, especially when it came to analysing the data and telling you something meaningful about the part, like whether it was in spec or not....

We ended up scanning our parts at 3600 points per revolution (0.1 degree increments) at 3 elevations and post processed the data, including centering the data in the tolerance zone, and statistically analysing all 100000 points within less than 15 seconds. We used a C code application which we wrote ourselves to do the analysis. We gave 3 different companies a go at writing the code and it was a total waste of time and money...

Good luck with your application, its not a straightforward one...