Summary
Multispectral images can help monitor crops by showing light beyond what our eyes see, but these images are usually fuzzy and don’t show small details clearly. The authors tested special computer programs designed to sharpen regular color images to see if they work on multispectral crop images without messing up color details. They found that some newer programs called transformer-based methods do a much better job than others at making the images clearer while keeping important information accurate. This means farmers and scientists could get clearer images of crops to better understand plant health and traits.
What this means in practice
- •For agriculture monitoring teams: Improve spatial detail of multispectral crop images to enhance accuracy of vegetation health assessments without loss of spectral information.
- •For remote sensing software developers: Incorporate transformer-based super-resolution techniques trained on RGB images to enhance multispectral imagery resolution for crop analysis tools.
Authors
Abhiram Karukayil, João F. C. Mota, Mateo Vásquez-Guevara, Tito Arevalo-Ramirez, Fernando Auat Cheein
Abstract
Multispectral imagery provides spectral information beyond the visible range for vegetation monitoring, but it typically suffers from limited spatial resolution, which can obscure fine crop structures and affect vegetation-index (VI) estimates.} This study evaluates whether single-image super-resolution (SISR) methods developed for red, green, and blue (RGB) imagery can enhance the spatial resolution of red, green, and near-infrared (RGN) crop images while preserving spectral fidelity. Eight pretrained methods are applied directly to RGN images without fine-tuning: bicubic and nearest-neighbor interpolation, Real-ESRGAN, HAT, HiT-SR, SRFormer, ResShift, and StableSR. Reconstruction at $4\times$ and $16\times$ is evaluated for whole images and manually segmented leaf regions. Reconstruction quality is assessed using mean absolute error (MAE), peak signal-to-noise ratio (PSNR), structural similarity index (SSIM), spectral angle mapper (SAM), and erreur relative globale adimensionnelle de synthèse (ERGAS). Eight VIs, computed from reference and super-resolved images, are evaluated using MAE, PSNR, and SSIM. Transformer-based methods \modification{significantly} outperform GAN- and diffusion-based approaches in both evaluation settings \modification{($p < 0.001$)}, with \modification{HAT, HiT-SR, and SRFormer forming the highest-performing statistical tier}. These findings indicate that RGB-trained SISR can enhance spatial detail in RGN crop imagery while maintaining spectral consistency and VI reliability for high-resolution crop phenotyping.