| Title |
Orientacijos sistemos, naudojančios poliarizacinius filtrus, tyrimas / |
| Translation of Title |
Research of orientation system using polarizing filters. |
| Authors |
Garbašauskas, Matas |
| Full Text |
|
| Pages |
67 |
| Keywords [eng] |
road segmentation ; computer vision ; polarizing filters ; image processing |
| Abstract [eng] |
In practice robot navigation can be done by processing information acquired from front mounted video camera. Usually the sample information is used for road segmentation and recognition. To be more specific we can take a small video sample and extract important information that can be used to compare and recognize the road pattern. This road segmentation concept is utilized in the Gaussian mixture model, template matching and Haar features algorithms. Although these algorithms are effective in some cases even slight changes in road color may cause serious misclassifications. In this work our goal is to research and develop a road segmentation system which work would not be effected by distance from sample zone and test if polarizing filters would improve performance of this system. In addition to developing this system we test its recognition result in various situations: 1. different sample area; 2. shadows and road texture imperfections; 3. poor visibility; 4. various road textures; 5. adding polarizing filters. |
| Dissertation Institution |
Kauno technologijos universitetas. |
| Type |
Master thesis |
| Language |
Lithuanian |
| Publication date |
2015 |
