EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

AMRF-adaptive multi resolution fusion for enhanced salient object detection using deep learning

Mueen Uddin, Maha Abdelhaq, Meshari Huwaytim Alanazi, Sedat Akleylek, Korhan Cengiz, Raed Alsaqour

PeerJ Computer Science · 23.07.2026 · 12 · e4027

DOI: 10.7717/peerj-cs.4027 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PeerJ Computer Science, online erschienen am 23.07.2026. Verfasst von Mueen Uddin und 5 weiteren Autor:innen. Die Arbeit nennt 56 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia, Northern Border University, Arar, KSA.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Salient object detection (SOD) aims to focus the most prominent objects in images for many vision applications such as object tracking, image and video segmentation and also some medical image analysis tasks like tumor highlighting. Recent methods have some limitations in which most of existing approaches are designed for handling images of fixed sizes and secondly, they often fail to capture objects in complicated backgrounds. In this article, we present an Adaptive Multi-Resolution Fusion (AMRF) approach for overcoming these limitations. The approach, called as AMRF, consists of several deep learning networks, which exploit effective features at different resolutions. The AMRF framework extracts edge-preserving features using convolutional neural networks (CNNs) and encodes contextual dependencies through gated recurrent neural networks (RNNs), followed by image decomposition into full, half, and quarter resolutions. A softmax-driven energy model learns per-image scale weights γn, with an attention-guided fusion mechanism producing the fused feature map. Saliency maps are generated via a U-Net architecture optimized by a combined binary cross-entropy (BCE) and Intersection over Union (IoU) loss, while adversarial refinement using generative adversarial networks (GANs) sharpens boundary precision. An adaptive regulator dynamically adjusts the weighting parameters (α, β) to balance local and global evidence under the composite loss L total . Experimental evaluation on MSRA-B, DUTS, and HKU-IS datasets demonstrates that AMRF achieves 95.1% precision, 92.8% recall, 93.9% F-measure, and a mean absolute error (MAE) of 0.056.

Förderung

  • Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia
  • Northern Border University, Arar, KSA

Zitieren

APA

Uddin, M., Abdelhaq, M., Alanazi, M. H., Akleylek, S., Cengiz, K., & Alsaqour, R. (2026). AMRF-adaptive multi resolution fusion for enhanced salient object detection using deep learning. PeerJ Computer Science, 12, e4027. https://doi.org/10.7717/peerj-cs.4027

Vancouver

Uddin M, Abdelhaq M, Alanazi MH, Akleylek S, Cengiz K, Alsaqour R. AMRF-adaptive multi resolution fusion for enhanced salient object detection using deep learning. PeerJ Computer Science. 2026;12:e4027. doi:10.7717/peerj-cs.4027

BibTeX
@article{Uddin2026,
  author = {Uddin, Mueen and Abdelhaq, Maha and Alanazi, Meshari Huwaytim and Akleylek, Sedat and Cengiz, Korhan and Alsaqour, Raed},
  title = {AMRF-adaptive multi resolution fusion for enhanced salient object detection using deep learning},
  journal = {PeerJ Computer Science},
  year = {2026},
  volume = {12},
  pages = {e4027},
  doi = {10.7717/peerj-cs.4027}
}
Kostenlos starten →