Open-set remote sensing object detection method based on latent-space structural regularization

Qiying Ling

doi:10.1016/j.eswa.2026.134395

Open-set remote sensing object detection method based on latent-space structural regularization

2026-09-12

0

PRE

AI

DOI：10.1016/j.eswa.2026.134395

Original

Original request for help

Save

Abstract

En 中文

• An open-set remote sensing object detection method with latent-space diffusion regularization. • Latent space contrastive regularization enables unknown samples to improve discrimination. • Order-preserving diffusion representation learning reduces over-clustering of unknown objects. • The proposed method improves discrimination of known and unknown objects and generalization.

Keywords:

Open-set detection

Remote sensing images

Structural regularization

Open-set remote sensing object detection method based on latent-space structural regularization

Abstract

Journal

Organization

Cited Papers