UDHF2-Net: Uncertainty-diffusion-model-based High-Frequency TransFormer Network for Remotely Sensed Imagery Interpretation

TL;DR

This paper introduces UDHF2-Net, a novel high-frequency transformer network that enhances remote sensing imagery interpretation by improving edge detection, reducing false alarms, and effectively handling spectrum noise through innovative modules and a semi-pseudo-Siamese architecture.

Contribution

The paper proposes the first uncertainty-diffusion-model-based high-frequency transformer network with novel modules for improved edge extraction and change detection in remote sensing imagery.

Findings

Enhanced edge extraction accuracy through high-frequency streams.

Reduced false change detection caused by geometric registration errors.

Effective noise removal using self-supervised uncertainty diffusion module.

Abstract

Remotely sensed imagery interpretation (RSII) faces the three major problems: (1) objective representation of spatial distribution patterns; (2) edge uncertainty problem caused by downsampling encoder and intrinsic edge noises (e.g., mixed pixel and edge occlusion etc.); and (3) false detection problem caused by geometric registration error in change detection. To solve the aforementioned problems, uncertainty-diffusion-model-based high-Frequency TransFormer network (UDHF2-Net) is the first to be proposed, whose superiorities are as follows: (1) a spatially-stationary-and-non-stationary high-frequency connection paradigm (SHCP) is proposed to enhance the interaction of spatially frequency-wise stationary and non-stationary features to yield high-fidelity edge extraction result. Inspired by HRFormer, SHCP proposes high-frequency-wise stream to replace high-resolution-wise stream in HRFormer through the whole encoder-decoder process with parallel frequency-wise high-to-low streams, so it improves the edge extraction accuracy by continuously remaining high-frequency information; (2) a mask-and-geo-knowledge-based uncertainty diffusion module (MUDM), which is a self-supervised learning strategy, is proposed to improve the edge accuracy of extraction and change detection by gradually removing the simulated spectrum noises based on geo-knowledge and the generated diffused spectrum noises; (3) a frequency-wise semi-pseudo-Siamese UDHF2-Net is the first to be proposed to balance accuracy and complexity for change detection. Besides the aforementioned spectrum noises in semantic segmentation, MUDM is also a self-supervised learning strategy to effectively reduce the edge false change detection from the generated imagery with geometric registration error.