TRNet: Topography-Guided Frequency Rectification and Structure-Aware Decoding for Multimodal Paddy Rice Segmentation

2026-08-04

AI Summary

This study addresses the challenge of accurately segmenting rice paddies in high-resolution remote sensing imagery over mountainous and hilly regions, where terrain-induced spectral distortions often cause confusion with spectrally similar vegetation. To mitigate this issue, the authors propose a novel framework that integrates 0.5-meter RGB imagery with 5-meter digital elevation model (DEM) and slope data. A terrain energy–spectral correction module is introduced to perform low-frequency modulation and high-frequency regulation at an early stage, followed by a terrain-guided structure-aware decoder that jointly optimizes semantic, boundary, and internal cues. The method effectively suppresses interference from steep slopes while enhancing rice features in flatter areas. Evaluated on test sets Area A and B, the approach achieves rice IoU scores of 85.10% and 80.68%, respectively—improvements of 9.15 and 18.83 percentage points over the baseline—and substantially reduces false detection rates.

Abstract

Mapping paddy rice from very-high-resolution imagery in mountainous and hilly regions is difficult because terrain alters optical appearance and increases confusion with visually similar vegetation. We present TRNet for 0.5-m GaoJing-1 red--green--blue (RGB) imagery, a 5-m TanDEM-X digital elevation model (DEM), and derived slope. Separate visual and terrain encoders preserve modality-specific features. At an early encoder stage, Topographic Energy-Spectral Rectification applies terrain-conditioned low-frequency modulation and asymmetric high-frequency regulation to suppress steep-slope clutter and conditionally enhance compatible low-slope rice cues. The Topography-guided Paddy Structure Decoder combines semantic, rice--background boundary, and interior cues, using coarse terrain as context. Experiments used an Area A internal test set and held-out Area B, which had steeper terrain and lower rice prevalence. TRNet achieved rice intersection-over-union (IoU) values of 85.10\% and 80.68\%, exceeding the original Dual-Encoder U-Net by 9.15 and 18.83 percentage points, respectively. Ablation and slope-stratified results linked these gains to frequency rectification, structure learning, and fewer steep-terrain false positives. The results support coarse topography as a contextual prior for very-high-resolution paddy rice mapping.

Problem

• paddy rice segmentation
• mountainous regions
• terrain effects
• very-high-resolution imagery
• vegetation confusion

Innovation

• Topography-Guided Frequency Rectification
• Structure-Aware Decoding
• Multimodal Segmentation
• Paddy Rice Mapping
• Digital Elevation Model