SOLO-Net: A Sparser but Wiser Method for Small Object Detection in Remote-Sensing Images

Small object detection in remote-sensing images is essential, yet challenging due to the unique characteristics of small objects. On the one hand, it can be difficult to distinguish small objects from complex backgrounds. On the other hand, due to their minuscule size, small objects are also easily...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:IEEE geoscience and remote sensing letters Ročník 21; s. 1 - 5
Hlavní autoři: Lei, Mengqi, Liu, Xianyuan
Médium: Journal Article
Jazyk:angličtina
Vydáno: Piscataway IEEE 2024
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Témata:
Aggregation
Convergence
Detection
Feature extraction
Modules
Object detection
Object recognition
Performance enhancement
Performance evaluation
Remote sensing
Remote-sensing images
Sensors
Sigmoid-intersection over union (IoU) loss
small object detection
Sports
top-k sparse outlook (TKSO) attention
Training
ISSN:1545-598X, 1558-0571
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Small object detection in remote-sensing images is essential, yet challenging due to the unique characteristics of small objects. On the one hand, it can be difficult to distinguish small objects from complex backgrounds. On the other hand, due to their minuscule size, small objects are also easily submerged during feature fusion. This letter proposes a novel detection method called sparse outlook network (SOLO-Net) to address these issues. First, we propose a top-k sparse outlook (TKSO) attention module and the sparse outlook path aggregation network (SOLO-PAN) as the fundamental component of SOLO-Net. This module improves the performance of the path aggregation network (PANet), thereby enhancing the ability of the model to focus on small objects. Second, we propose a Sigmoid-intersection over union (IoU) loss function specifically designed for small objects, accelerating model convergence and improving detection performance. Finally, we evaluate our model on the RSOD and DIOR datasets, achieving mean average precision (mAP) scores of 93.8% and 74.5%, respectively.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1545-598X
1558-0571
DOI:10.1109/LGRS.2023.3337807