Despite significant advancements in salient object detection (SOD) in optical remote sensing images (ORSIs), challenges persist due to the intricate edge structures of ORSIs and the complexity of their contextual relationships. Current deep learning (DL) approaches encounter difficulties in accurately identifying boundary features and lack efficiency in collaboratively modeling the foreground and background by leveraging contextual features. To address these challenges, we propose a stronger multifaceted collaborative salient object detector in ORSIs, termed LBA-MCNet, which incorporates aspects of localization, balance, and affinity. The network focuses on accurately locating targets, balancing detailed features, and modeling image-level global context information. Specifically, we design the edge feature adaptive balancing and adjusting (EFABA) module for precise edge localization, using edge features to guide attention to boundaries and preserve spatial details. Moreover, we design the global distributed affinity learning (GDAL) module to model global context. It captures global context by generating an affinity map from the encoder’s final layer, ensuring effective modeling of global patterns. In addition, deep supervision during deconvolution further enhances feature representation. Finally, we compared with 28 state-of-the-art approaches on three publicly available datasets. The results clearly demonstrate the superiority of our method. The codes of our method are available at https://github.com/little1bold/LBA-MCNet .
22 August 2024 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Deep learning methods are widely used in building information extraction from remote sensing images (RSIs). However, this task still faces great challenges. Firstly, it is difficult to perform accurate boundary localization due to the complex contextual relationship of the building boundary area; Secondly, the heterogeneity caused by different materials on building tops, as well as environmental factors such as climate and vegetation, complicates the extraction of top information; Lastly, given the complexity of the RSIs, the accuracy and generalization of the existing method still need to be further enhanced. This study introduces the Boundary Enhancement and Multi-scale Refinement Fusion Network (BEMRF-Net) to overcome these challenges. The Boundary-aware Self-Attention Module (BSA) is initially proposed to refine the network's boundary detection capabilities. It applies the distance transform algorithm across both channel and spatial dimensions, reducing boundary fluctuation and enhancing the accuracy of building boundary extraction. Subsequently, we present the Multi-Scale Refined Fusion Module (MSRF) to resolve discontinuities within buildings caused by rooftop obstructions. This module effectively merges high and low-level features, overcoming information gaps through the strategic integration of Multi-scale data. To demonstrate the efficacy of our methodology, we conducted comprehensive experiments and analyses using the WHU Building Dataset and the Inria Aerial Image Labeling Dataset, both known for their complexity. Our method surpassed 16 contemporary state-of-the-art (SOTA) techniques, achieving IoU scores of 91.15% and 79.52% for each dataset, respectively, marking the highest accuracy levels reported. Further, in-depth discussions on efficiency, generalizability, and ablation studies emphasized our method's robustness and adaptability.
13 November 2023 in IEEE Transactions on Geoscience and Remote Sensing
With the increasing application of deep learning in various domains, salient object detection in optical remote sensing images (ORSI-SOD) has attracted significant attention. However, most existing ORSI-SOD methods predominantly rely on local information from low-level features to infer salient boundary cues and supervise them using boundary ground truth, but fail to sufficiently optimize and protect the local information, and almost all approaches ignore the potential advantages offered by the last layer of the decoder to maintain the integrity of saliency maps. To address these issues, we propose a novel method named boundary-semantic collaborative guidance network (BSCGNet) with dual-stream feedback mechanism. First, we propose a boundary protection calibration (BPC) module, which effectively reduces the loss of edge position information during forward propagation and suppresses noise in low-level features without relying on boundary ground truth. Second, based on the BPC module, a dual feature feedback complementary (DFFC) module is proposed, which aggregates boundary-semantic dual features and provides effective feedback to coordinate features across different layers, thereby enhancing cross-scale knowledge communication. Finally, to obtain more complete saliency maps, we consider the uniqueness of the last layer of the decoder for the first time and propose the adaptive feedback refinement (AFR) module, which further refines feature representation and eliminates differences between features through a unique feedback mechanism. Extensive experiments on three benchmark datasets demonstrate that BSCGNet exhibits distinct advantages in challenging scenarios and outperforms the 17 state-of-the-art (SOTA) approaches proposed in recent years. Codes and results have been released on GitHub: https://github.com/YUHsss/BSCGNet.