Találati lista | Tudóstér

001-es BibID:	BIBFORM076396
035-os BibID:	(Scopus)85062828739
Első szerző:	Abriha Dávid (geográfus)
Cím:	Identification of roofing materials with Discriminant Function Analysis and Random Forest classifiers on pan-sharpened WorldView-2 imagery - a comparison / Abriha Dávid, Kovács Zoltán, Ninsawat Sarawut, Bertalan László, Balázs Boglárka, Szabó Szilárd
Dátum:	2018
ISSN:	2064-5031 2064-5147
Tárgyszavak:	Természettudományok Földtudományok idegen nyelvű folyóiratközlemény hazai lapban folyóiratcikk REMOTE SENSING pan-sharpening asbestos MACHINE LEARNING
Megjelenés:	Hungarian Geographical Bulletin. - 67 : 4 (2018), p. 375-392. -
További szerzők:	Kovács Zoltán (1988-) (geográfus) Ninsawat, Sarawut Bertalan László (1989-) (geográfus) Balázs Boglárka (1985-) (geográfus) Szabó Szilárd (1974-) (geográfus)
Pályázati támogatás:	EFOP-3.6.1-16-2016-00022 EFOP TNN 123457 Egyéb ÚNKP-18-2-I-DE-81 Egyéb
Internet cím:	Szerző által megadott URL DOI Intézményi repozitóriumban (DEA) tárolt változat
Borító:
Saját polcon:

001-es BibID:	BIBFORM132954
Első szerző:	Szabó Szilárd (geográfus)
Cím:	Lithological mapping with pseudo-labeling: promise or overestimation in data-scarce settings? / Szilárd Szabó, Abdelmajeed A. Elrasheed, Lilla Kovács, Imre J. Holb, Szilárd Likó, Dávid Abriha
Dátum:	2025
ISSN:	2064-5031 2064-5147
Megjegyzések:	Reference data are the most crucial points in model building. In geoscience, a scarcity of sufficient reference data is common. Pseudo-labelling (PL), i.e., incorporating high-probability data in the model building process, offers a potential solution. We aimed to reveal the efficiency of PL in lithological mapping in a vegetation-free arid region of Sudan. Random Forest (RF) and Multiple Adaptive Regression Splines (MARS) were used to classify a Landsat 9 image. Reference data were collected during field work and visual interpretation. Image processing yielded classified maps with associated probability layers, from which 1000 additional traditional samples (PL data) were extracted at a 95% probability. A detailed accuracy assessment was conducted, and accuracy measures were evaluated using statistical analysis and visual inspection. MARS was found to be an ambiguous classifier because the probability was too optimistic related to the overall accuracy (OA) (81% of samples had above 99% probability, OA=98.2%) compared to RF (21% above 99%, OA=98.1%); that is, despite the high probability, the accuracy improvement was only 0.1%. At the class level, the correlation between probability and the F1-score was low (0.21%). The original and PL-based models resulted in different maps with improved accuracy, although the new model version showed lower probability values for both the classifiers. Visual inspection proved essential for better insights into the spatial patterns: expert knowledge is crucial for controlling the occurrence of rock types and identifying false classifications. The main finding is that probability should be handled carefully, as it does not guarantee high model performance in classification, although the PL approach can lead to more reliable maps
Tárgyszavak:	Természettudományok Földtudományok idegen nyelvű folyóiratközlemény hazai lapban folyóiratcikk Random Forest Multiple Adaptive Regression Splines self-training probability data augmentation
Megjelenés:	Hungarian Geographical Bulletin. - "Accepted by Publisher" (2025), p. 1-29. -
További szerzők:	Abdelmajeed, Adam Elrasheed Ali (1988-) (Geologist) Kovács Lilla (Msc hallgató) Holb Imre (1973-) (agrármérnök) Likó Szilárd Balázs Abriha Dávid (1995-) (geográfus)
Pályázati támogatás:	TÁMOP-4.1.1.C-13/1/KONV-2014-0001 SUPPORT
Internet cím:	Intézményi repozitóriumban (DEA) tárolt változat
Borító:
Saját polcon: