Effects of Atmospheric Correction and Image Enhancement on Effective Plastic Greenhouse Segments Based on a Semi-Automatic Extraction Method

To improve the multi-resolution segmentation (MRS) quality of plastic greenhouses (PGs) in GaoFen-2 (GF-2) images, effects atmospheric correction and image enhancement on effective PG segments (EPGSs) were evaluated. A new semi-automatic method was also proposed to extract EPGSs an accurate efficient way. Firstly, GF-2 images preprocessed via correction, orthographical registration, fusion, linear compression, or spatial filtering, and, then, boundary-removed point samples with adjustable density made based reference polygons by taking advantage characteristics chessboard segmentation. Subsequently, used quickly accurately containing 70% greater pixels each MRS result. Finally, extracted compared analyzed intersection over union (IoU), over-segmentation index (OSI), under-segmentation (USI), error total area (ETA), composite (CEI). The experimental results show that, along change control variables, optimal scale parameter, time segmentation, IoU, OSI, USI, CEI all showed strong changing trends, values ETA close 0. Furthermore, group, CEIs from those corrected enhanced resulted lower values, involved linearly compressing DN value atmospheric-corrected fusion 0–255, then using Fast Fourier Transform a circular low-pass filter radius 800 frequency domain; this case, had minimum 0.159. study indicate that design experiment is reasonable pixel ratio determine EPGSs, OSI-USI-ETA-CEI pattern can be more than IoU when it needed evaluate EPGSs. Moreover, into consideration heterogeneity target characteristics, prior