Optimization of the Modelling Workflow of a 3D Scanner

OData support
Supervisor:
Dr. Kovács Tibor
Department of Automation and Applied Informatics

My first task was to learn and understand the department's laser scanner. Learning about a variety of 3D scanning technologies and all I could about the department's scanner, positive and negative features as well, helped to assess many points of development. The analysis of the previous interface from an ergonomic point of view was essential to its transformation to a better, faster and easier to use interface.

Changes to the current user interface consisted mainly in removing redundant and confusing elements, and bringing functions that are of use to the user to the main window. The new procedure’s surface is similar to installation wizards and it is easily accessible to users by a press of a button, and it can help a user who has never used the laser scanner before. When using the wizard modifying the settings interface is not designed for the average user, but more for the ones who are studying or working in the information technology. The main reason is that the 3D scanning process includes decisions that can only be made by people who understand the limits of the hardware and software.

The tool that I have created for the purpose of refining the scanned objects is a subsystem, which can be found in other 3D and image editing programs and should be familiar to users. One of the reasons to do this was to make it easier for the user to find and use this subsystem.

The subsystem itself is actually a pipeline, which is getting the raw scan of the object at the beginning, then moving through the pipe while passing through different filters the object is converted to the optimized model by the end. I also had to pay attention that the raw scan does not match the object type of exported objects, and between these types, the program has also had to perform a conversion, so I had to create two different types of filter accordingly. One of them can be used to edit the raw model, while the other one is used for the transformed one. There are advantages for both types of filter to use, and most of them are not compatible to the other type, so it was necessarry to develop both types of filters.

The users while using the subsystem can use two stacks, each storing its own kind of filters. The order, properties and the visibility of the filters can be changed using the graphical user interface. In addition, I defined an interface that will provide a good template for developers in the future, who want to add more filters to the program. In this semester I already created a few useful filter, both for the raw and the transformed model. I tried to create the filters so that their processing time is as optimized as possible.

My last task was to test the different filters in the subsystem and examine the results. Because the object becoming more beautiful can not be measured from an objective point of view, therefore I mainly measured processing time and storage usage of the filters, but of course, I checked independently in each case, if the filter fulfilled it’s task, and if it can really help the user in creating a better and more optimized model.

Downloads

Please sign in to download the files of this thesis.