1. Region segmentation of images based on a raster-scan paradigmLuka Lukač, Andrej Nerat, Damjan Strnad, Štefan Horvat, Borut Žalik, 2024, original scientific article Abstract: This paper introduces a new method for the region segmentation of images. The approach is based on the raster-scan paradigm and builds the segments incrementally. The pixels are processed in the raster-scan order, while the construction of the segments is based on a distance metric in regard to the already segmented pixels in the neighbourhood. The segmentation procedure operates in linear time according to the total number of pixels. The proposed method, named the RSM (raster-scan segmentation method), was tested on selected images from the popular benchmark datasets MS COCO and DIV2K. The experimental results indicate that our method successfully extracts regions with similar pixel values. Furthermore, a comparison with two of the well-known segmentation methods—Watershed and DBSCAN—demonstrates that the proposed approach is superior in regard to efficiency while yielding visually similar results.
Keywords: segment, image analysis, distance metric, Watershed, DBSCAN
Published in DKUM: 05.12.2024; Views: 0; Downloads: 2
 Link to file |
2. Detection and Monitoring of Woody Vegetation Landscape Features Using Periodic Aerial PhotographyDamjan Strnad, Štefan Horvat, Domen Mongus, Danijel Ivajnšič, Štefan Kohek, 2023, original scientific article Keywords: woody vegetation landscape features, change detection, segmentation neural network, cyclic aerial photography, digital orthophoto
Published in DKUM: 23.05.2024; Views: 196; Downloads: 15
 Full text (6,12 MB)
This document has many files! More... |
3. Napovedovanje dinamike plazu urbas z modeli časovnih vrst in strojnim učenjemŠtefan Horvat, 2022, master's thesis Abstract: Plazovi lahko resno ogrozijo človeška življenja in povzročijo ogromno gmotno škodo. Na dinamiko plazu običajno vpliva večje število zunanjih dejavnikov, zato je napovedovanje premikov težka naloga. V sodobnem času lahko premike plazov podrobno spremljamo z natančnimi merilnimi instrumenti in tako tvorimo množico podatkov, na podlagi katere gradimo razlagalne in napovedne modele. V magistrskem delu preizkušamo različne tehnike modeliranja premikov plazu Urbas, ki spada med bolj aktivne plazove v Sloveniji. Za modeliranje dinamike plazu uporabimo modele časovnih vrst in nevronsko mrežo z dolgim kratkoročnim spominom. Najboljše prileganje je dosegla nevronska mreža z dolgin kratkoročnim spominom, katere srednja kvadratna napaka je znašala 3,37 mm. Pri napovedovanju premikov se je najbolje odrezal linearni regresijski model s srednjo kvadratno napako 0,52 mm.
Keywords: plaz, časovne vrste, linearna regresija, dinamična regresija, nevron-ske mreže LSTM
Published in DKUM: 15.12.2022; Views: 942; Downloads: 153
Full text (4,33 MB) |
4. |
