| Title |
Dirbtinio intelekto metodų taikymas įvairių defektų atpažinimui / |
| Translation of Title |
Application of artificial intelligence methods for the recognition of different defects. |
| Authors |
Gudas, Lukas Artūras |
| Full Text |
|
| Pages |
55 |
| Keywords [eng] |
defects ; images analysis ; one-class ; classification |
| Abstract [eng] |
In this work similar area articles are reviewed. In the research part it is chosen to use one-class classification, because it helps to avoid the problem of gathering the library of defects, since there can be unforseen types of defects. Few decisions are made for preprocessing the images. In this thesis 11 feature extraction parameters are tried for use in image classification. According to the results the best classificator is chosen with the most suitable extracted features. |
| Dissertation Institution |
Kauno technologijos universitetas. |
| Type |
Master thesis |
| Language |
Lithuanian |
| Publication date |
2017 |
