Repository logo

Browsing by Subject "rocks classification"

Filter results by typing the first few letters
Now showing 1 - 1 of 1
  • thumbnail.journal.alt
    Item type:Thesis, Access status: Restricted ,
    Zastosowanie metod rozpoznawania obrazów do klasyfikacji wybranych typów skał
    (Data obrony: 2012-07-10) Górszczyk, Andrzej
    Wydział Geologii, Geofizyki i Ochrony Środowiska
    The dynamic development of artificial intelligence is reflected by the increased use of its techniques. Research and projects in areas such as medicine, robotics, biometrics, military, diagnostic systems and many others for the most part now rely on automated processes, which are based on computer pattern recognition. This trend is also confirmed for studies related to the wider geology, where in conjunction with image processing algorithms and image analysis, pattern recognition is a useful tool for classification and description of rock structures. This paper shows how these techniques can be used to identify some types of rocks based on their microscopic images. The author compiled the work to compare the effectiveness of five methods of pattern recognition based on each of the nine selected rocks, whose images were processed in four color spaces. The best summary of the method of NN and CIELab color space gave a result equal to the correct classification of 99.81% which demonstrates the effectiveness of pattern recognition to the shown issue and can lead to work on more complex problems of a similar scope.