Deeply Hybrid Contrastive Learning Based on Semantic Pseudo-Label for Salient Object Detection in Optical Remote Sensing Images
- 14 June 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 26 (15209210) , 10892-10907
Abstract
Salient object detection in natural scene images (NSI-SOD) has undergone remarkable advancements in recent years. However, compared to those of natural images, the properties of remote sensing images (ORSIs), such as diverse spatial resolutions, complex background structures, and varying visual attributes of objects, are more complicated. Hence, how to explore the multiscale structural perceptual information of ORSIs to accurately detect salient objects is more challenging. In this paper, inspired by the superiority of contrastive learning, we propose a novel training paradigm for ORSI-SOD, named Deeply Hybrid Contrastive Learning Based on Semantic Pseudo-Label (DHCont), to force the network to extract rich structural perceptual information and further learn the better-structured feature embedding spaces. Specifically, DHCont first splits the ORSI into several local subregions composed of color- and texture-similar pixels, which act as semantic pseudo-labels. This strategy can effectively explore the underdeveloped semantic categories in ORSI-SOD. To delve deeper into multiscale structure-aware optimization, DHCont incorporates a hybrid contrast strategy that integrates “pixel-to-pixel”, “region-to-region”, “pixel-to-region”, and “region-to-pixel” contrasts at multiple scales. Additionally, to enhance the edge details of salient regions, we develop a hard edge contrast strategy that focuses on improving the detection accuracy of hard pixels near the object boundary. Moreover, we introduce a deep contrast algorithm that adds additional deep-level constraints to the feature spaces of multiple stages. Extensive experiments on two popular ORSI-SOD datasets demonstrate that simply integrating our DHCont into the existing ORSI-SOD models can significantly improve the performance.Keywords
Funding Information
- China Postdoctoral Science Foundation (2023M731803, 2023M741126)
- National Natural Science Foundation of China (62303173)
- China National Postdoctoral Program for Innovative Talents (BX20230108)
This publication has 76 references indexed in Scilit:
- Position-based anchor optimization for point supervised dense nuclei detectionNeural Networks, 2023
- Deep unsupervised part-whole relational visual saliencyNeurocomputing, 2023
- Local aggregation and global attention network for hyperspectral image classification with spectral-induced aligned superpixel segmentationExpert Systems with Applications, 2023
- Delving into Universal Lesion Segmentation: Method, Dataset, and Benchmark2022
- A parallel down-up fusion network for salient object detection in optical remote sensing imagesNeurocomputing, 2020
- U2-Net: Going deeper with nested U-structure for salient object detectionPattern Recognition, 2020
- Adaptive Affinity Fields for Semantic Segmentation2018
- SEEDS: Superpixels Extracted via Energy-Driven Sampling2012
- Quick Shift and Kernel Methods for Mode Seeking2008
- Efficient Graph-Based Image SegmentationInternational Journal of Computer Vision, 2004