Few Shot Transferability of Spatial Deep Learning for Crop Mapping using SAR Polarimetry in Highly Fragmented Farmlands

Jun-Wei Guo

Presentation information

[E] Oral

H (Human Geosciences ) » H-TT Technology & Techniques

[H-TT15] Geographic Information Systems and Cartography

Wed. May 27, 2026 9:00 AM - 10:30 AM Exhibition Hall Special Setting (2) (Exhibition Hall 7&8, Makuhari Messe)

Chairperson:Oguchi Takashi(Center for Spatial Information Science, The University of Tokyo), Wang Ruci(Center for Environmrntal Remote Sensing, Chiba University)

9:15 AM - 9:30 AM

[HTT15-02] Few Shot Transferability of Spatial Deep Learning for Crop Mapping using SAR Polarimetry in Highly Fragmented Farmlands

★Invited Papers

*Jun-Wei Guo¹, Yunung Nina Lin², Jen-Yu Han¹, Heather McNairn³, Xianfeng Jiao³, Andrew Davidson³, Li-Yu Liu⁴, Yi-Ting Zhang⁵ (1. Dept. Civil Eng., National Taiwan Univ., 2. Inst. Earth Sciences, Academia Sinica, 3. ACGEO-AER, AAFC, Canada, 4. Dept. Agronomy, National Taiwan Univ., 5. TARI, Ministry of Agriculture, Taiwan)

Keywords:Crop Classification, Farmland Fragmentation, Polarimetric Synthetic Aperture Radar (PolSAR), Spatial Context, Deep Learning, Mixed Pixels