AOSR-Net: All-in-One Sandstorm Removal Network

The framework of the proposed algorithm

Abstract

Most existing sandstorm image enhancement methods are based on traditional theory and prior knowledge, which often limit their generalization ability in real-world applications. In addition, these approaches often adopt a strategy of color correction followed by dust removal, which makes the algorithm structure too complex. To solve the issue, we propose a novel image restoration model, named all-in-one sandstorm removal network (AOSR-Net). It is designed based on a re-formulated sandstorm scattering model, which directly establishes the image mapping relationship by integrating intermediate parameters. Such integration strategy can effectively solve the issues of over-enhancement and weak generalization in the field of sand dust image enhancement. Experimental results on both synthetic and real-world sandstorm images illustrate that the proposed AOSR-Net is superior than the state-of-the-art methods.

Type
Publication
In 2023 IEEE 35th International Conference on Tools with Artificial Intelligence
Click the Cite button above to demo the feature to enable visitors to import publication metadata into their reference management software.
Yazhong Si
Yazhong Si
Hebei University of Technology
Ning Cheng
Ning Cheng
Director