AI Paper Atlas

Research map / Image generation / Image restoration

Multi-Window Gabor Transform Network for Ground Penetrating Radar B-Scan Image Reconstruction

Huabin Wang, Yu Yang, Xinran Zhong, Zilong Ling

AAAI 2026 · Image generation › Image restoration · approach: Reflection removal · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add defect guidance network and guidance loss to baseline (ablation)U-Net baseline · GRD (GPR B-scan reconstruction)SSIM (%): 67.93 -> 71.05Improved
add defect guidance network and guidance loss to baseline (ablation)U-Net baseline · GRD (GPR B-scan reconstruction)PSNR: 28.02 -> 29.29Improved
add single-window Gabor transform module to baseline (ablation)U-Net baseline · GRD (GPR B-scan reconstruction)SSIM (%): 67.93 -> 75.06Improved
replace single-window with multi-window Gabor module (ablation)U-Net baseline · GRD (GPR B-scan reconstruction)SSIM (%): 67.93 -> 78.59Improved
replace single-window with multi-window Gabor module (ablation)U-Net baseline · GRD (GPR B-scan reconstruction)PSNR: 29.68 -> 30.21Improved
add defect guidance network to multi-window Gabor model (ablation)MGT-Net · GRD (GPR B-scan reconstruction)SSIM (%): 78.59 -> 81.72Improved
add defect guidance network to multi-window Gabor model (ablation)MGT-Net · GRD (GPR B-scan reconstruction)PSNR: 30.21 -> 30.50Improved
set pixel-loss weight lambda1 to 10 instead of 5 (ablation)MGT-Net · GRD (GPR B-scan reconstruction)SSIM (%): 80.99 -> 81.72Improved
set guidance-loss weight lambda2 to 1 instead of 20 (ablation)MGT-Net · GRD (GPR B-scan reconstruction)SSIM (%): 78.83 -> 81.72Improved
raise pixel-loss weight lambda1 from 10 to 20 (ablation)MGT-Net · GRD (GPR B-scan reconstruction)PSNR: 30.50 -> 29.97Hurt
train detector on MGT-Net reconstructions instead of DN reconstructionsdefect detector · GRD defect detection on reconstructed imagesIOU: 65.04 -> 66.43Improved
train detector on MGT-Net reconstructions instead of DN reconstructionsdefect detector · GRD defect detection on reconstructed imagesBER: 9.39 -> 7.01Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy