Two‐stage local attention network for salient object detection in remote sensing images. Issue 3 (22nd November 2022)

• Record Type: Journal Article
• Authors: Lin, Qihui; Xia, Lurui; Li, Sen; Chen, Wanfeng
• Is Part Of: IET image processing. Volume 17:Issue 3(2023)
• Page Start: 849
• Page End: 861
• Publication Date: 2022-11-22
• Subjects: cascade decoder -- loss function -- salient object detection -- window self‐attention; Image processing -- Periodicals; 621.36705
• DOI: 10.1049/ipr2.12677
• Languages: English
• ISSNs: 1751-9659

Abstract

Abstract: With the development of earth observation satellites, it is very meaningful to study methods for intelligently processing remote sensing images (RSIs). As an important pretreatment technique, salient object detection (SOD) can produce accurate saliency predictions in natural scenes. However, due to the complex object types, high noise levels, and different imaging angles in RSIs, it is still difficult to generate accurate saliency predictions in such images. This paper proposes a SOD method for RSIs, namely, the two‐stage local attention network (TLANet). It uses two‐stage attention (TSA) to capture the relationships between the pixels of features at different scales and uses an adjacent semantic refinement (ASR) module to enhance the flow of information across scales. In addition, a cascade structure is introduced to the decoder so that more semantic features can guide the feature decoding process. Moreover, a new loss function is proposed to improve the prediction effects obtained for salient edges. Through comprehensive experiments conducted on two RSIs datasets, it is demonstrated that the TLANet is superior to the state‐of‐the‐art methods in both qualitative and quantitative comparisons. Abstract : This paper proposes an SOD method for RSIs, namely, the two‐stage local attention network (TLANet). It uses two‐stage attention (TSA) to capture the relationships between the pixels of features at different scales and uses an adjacent semantic refinement (ASR) module to enhance the flow of information across scales. In addition, we introduce a cascade structure to the decoder so that more semantic features can guide the feature decoding process. Moreover, a new loss function is proposed to improve the prediction effects obtained for salient edges. Through comprehensive experiments conducted on two RSIs datasets, we demonstrate that the TLANet is superior to the state‐of‐the‐art methods in both qualitative and quantitative comparisons.