EUVIMEDEUVIMED entdecken
Crossref

Multi-layer Feature Cascade Fusion Spiking Neural Network for Object Detection

Yongqiang Ma iD, Bailin Guo, Xuetao Zhang iD

International Journal of Neural Systems · 2025-11 · 35 (11)

DOI: 10.1142/s0129065725500637 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Neural Systems, online erschienen am 15.09.2025. Verfasst von Yongqiang Ma und 2 weiteren Autor:innen. Beteiligte Einrichtung: State Key Laboratory of Human-Machine Hybrid Augmented Intelligence, National Engineering Research Center of Visual Information and Applications, Ins…. Hauptthemen laut MEDLINE: Neural Networks, Computer, Action Potentials, Neurons, Models, Neurological. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 31 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Spiking Neural Networks (SNNs), as a biologically inspired computational model, have garnered significant attention in object detection and image classification due to their event-driven mechanism and low-power characteristics. However, in object detection tasks, the residual structures in conventional networks introduce nonspiking operations, posing a critical challenge for SNNs. To address this issue, we propose a multi-layer feature cascade fusion SNN (MFCF-SNN) for object detection. During feature extraction, our novel multi-level cascaded feature extraction module replaces residual connections with cascade operations, eliminating nonspiking computations while enhancing gradient propagation to deeper layers. For downsampling, we introduce a pooling-convolution module that combines max-pooling and spiking convolution, effectively preserving feature information and improving gradient flow. These two modules collectively ensure pure spike-based computation while facilitating deep network training, thereby enhancing detection accuracy. Experimental results on the PASCAL VOC 2012 and SSDD datasets demonstrate state-of-the-art performance, validating the effectiveness of our approach in advancing SNN-based object detection.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Neural Networks, ComputerAction PotentialsNeuronsModels, NeurologicalPattern Recognition, AutomatedHumans

Stichwörter der Autor:innen

Spiking neural networkfeature cascade fusionlayered optimizationobject detection

Autor:innen und Institutionen

  • State Key Laboratory of Human-Machine Hybrid Augmented Intelligence, National Engineering Research Center of Visual Information and Applications, Institute of Artificial Intelligence and Robotics, Xi'an Jiaotong Univers…

ORCID-Profile: Yongqiang Ma, Xuetao Zhang

Zitieren

APA

Ma, Y., Guo, B., & Zhang, X. (2025). Multi-layer Feature Cascade Fusion Spiking Neural Network for Object Detection. International Journal of Neural Systems, 35(11). https://doi.org/10.1142/s0129065725500637

Vancouver

Ma Y, Guo B, Zhang X. Multi-layer Feature Cascade Fusion Spiking Neural Network for Object Detection. International Journal of Neural Systems. 2025;35(11). doi:10.1142/s0129065725500637

BibTeX
@article{Ma2025,
  author = {Ma, Yongqiang and Guo, Bailin and Zhang, Xuetao},
  title = {Multi-layer Feature Cascade Fusion Spiking Neural Network for Object Detection},
  journal = {International Journal of Neural Systems},
  year = {2025},
  volume = {35},
  number = {11},
  doi = {10.1142/s0129065725500637}
}
Kostenlos starten →