数字农科院2.0

Effective variance attention-enhanced diffusion model for crop field aerial image super resolution

文献类型： 外文期刊

作者： Xiangyu Lu;Jianlin Zhang;Rui Yang;Qina Yang;Mengyuan Chen;Hongxing Xu;Pinjun Wan;Jiawen Guo;Fei Liu

作者机构：

关键词： Aerial imagery;Diffusion model;Super-resolution;Super-resolution relative fidelity index;Variance attention

期刊名称： ISPRS Journal of Photogrammetry and Remote Sensing

ISSN： 0924-2716

年卷期： 2024 年 218 卷

页码：

收录情况： SCIE（2024版） ; ; EI（2024版）

摘要： Image super-resolution (SR) can significantly improve the resolution and quality of aerial imagery. Emerging diffusion models (DM) have shown superior image generation capabilities through multistep refinement. To explore their effectiveness on high-resolution cropland aerial imagery SR, we first built the CropSR dataset, which includes 321,992 samples for self-supervised SR training and two real-matched SR datasets from high-low altitude orthomosaics and fixed-point photography (CropSR-OR/FP) for testing. Inspired by the observed trend of decreasing image variance with higher flight altitude, we developed the Variance-Average-Spatial Attention (VASA). The VASA demonstrated effectiveness across various types of SR models, and we further developed the Efficient VASA-enhanced Diffusion Model (EVADM). To comprehensively and consistently evaluate the quality of SR models, we introduced the Super-resolution Relative Fidelity Index (SRFI), which considers both structural and perceptual similarity. On the × 2 and × 4 real SR datasets, EVADM reduced Fréchet-Inception-Distance (FID) by 14.6 and 8.0, respectively, along with SRFI gains of 27 % and 6 % compared to the baselines. The superior generalization ability of EVADM was further validated using the open Agriculture-Vision dataset. Extensive downstream case studies have demonstrated the high practicality of our SR method, indicating a promising avenue for realistic aerial imagery enhancement and effective downstream applications. The code and dataset for testing are available at https://github.com/HobbitArmy/EVADM.

分类号：

  • 相关文献

[1]Edge-preserving light-field image super-resolution via feature affine transformation network. Zou, Rong,Dai, Wenjie,Bai, Shenghe,Mu, Senlin. 2024

[2]A point-supervised algorithm with multiscale semantic enhancement for counting multiple crop plants from aerial imagery. Huibin Li,Huaiyang Liu,Wenbo Wang,Haozhou Wang,Qiangyi Yu,Jianping Qian,Wenbin Wu,Yun Shi,Changxing Geng. 2025

[3]Leveraging Text Mining and Network Analysis for the Diffusion of Agricultural Science and Technology Policies in China. Xiaohe Liang,Yu Wu,Jiajia Liu,Jiayu Zhuang,Tong Yuan,Ying Chen,Lizhen Cui,Ailian Zhou,Jiajia Zhou,Tong Li. 2025

[4]DM_CorrMatch: a semi-supervised semantic segmentation framework for rapeseed flower coverage estimation using UAV imagery. Li, Jie,Zhu, Chengyong,Yang, Chenbo,Zheng, Quan,Wang, Binhui,Tu, Jingmin,Zhang, Qian,Liu, Sheng,Wang, Xinfa,Qiao, Jiangwei. 2025

作者其他论文 更多>>