National Repository of Grey Literature 2 records found  Search took 0.00 seconds. 
Realistic methods of generating landscape models through the Unity framework
HEJPLÍK, Daniel
This thesis deals with research and development of a tool capable of automated creation of a virtual environment. Automated terrain generation is a non-trivial problem, the solution of which requires the creation of sophisticated methods that provide the possibility of data parameterization for the needs of the target environment. On the basis of this data, the product (basic terrain) is then processed and its appearance improved to match as much as possible what the real world looks like. To understand these processes, it is necessary to understand how environment is represented in 3D graphics and, consequently, the ways of texturing it. Also for this reason, part of the work is an examination and description of how these procedures work and what are the possibilities of using them. Furthermore, the work examines possibilities for terrain generation from simple spaces based on the generation of random values (perlin noise) to the subsequent improvement of its details. These more complex methods are based on nature-inspired processes (hydraulic erosion).
Identifikace objektů v obraze se zaměřením na aplikace v dopravě
HEJPLÍK, Daniel
This bachelor´s thesis has two parts, theoretical and practical. Theoretical part focuses on research and comparsion of algorithms and principles used for object recognition in digital images. The result is evaluation of their suitability for usage in traffic. Main evaluation criterion is speed of individual methods and reliability of object identification. Practical part focuses on creating specific software tool in Java language, based on object recognition using known algorithms, primarily library OpenCV. The program will include its own graphical interface. Part of result will be documentation for administrators and end users.

Interested in being notified about new results for this query?
Subscribe to the RSS feed.