SAEN-BGS: Energy-efficient spiking autoencoder network for background subtraction

Background subtraction (BGS) is utilized to detect moving objects in a video and is commonly employed at the onset of object tracking and human recognition processes. Nevertheless, existing BGS techniques utilizing deep learning still encounter challenges with various background noises in videos, in...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Pattern recognition Ročník 169; s. 111792
Hlavní autori: Zhang, Zhixuan, Li, Xiao Peng, Liu, Qi
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Elsevier Ltd 01.01.2026
Predmet:
Autoencoder
Background subtraction
Deep learning
Energy efficiency
Spiking neural network
Deep learning
Energy efficiency
Background subtraction
Autoencoder
Spiking neural network
ISSN:0031-3203
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:Background subtraction (BGS) is utilized to detect moving objects in a video and is commonly employed at the onset of object tracking and human recognition processes. Nevertheless, existing BGS techniques utilizing deep learning still encounter challenges with various background noises in videos, including variations in lighting, shifts in camera angles, and disturbances like air turbulence or swaying trees. To address this problem, we design a spiking autoencoder network, termed SAEN-BGS, based on noise resilience and time-sequence sensitivity of spiking neural networks (SNNs) to enhance the separation of foreground and background. To eliminate unnecessary background noise and preserve the important foreground elements, we begin by creating the continuous spiking conv-and-dconv block, which serves as the fundamental building block for the decoder in SAEN-BGS. Moreover, in striving for enhanced energy efficiency, we introduce a novel self-distillation spiking supervised learning method grounded in ANN-to-SNN frameworks, resulting in decreased power consumption. In extensive experiments conducted on CDnet-2014 and DAVIS-2016 datasets, our approach demonstrates superior segmentation performance relative to other baseline methods, even when challenged by complex scenarios with dynamic backgrounds. [Display omitted] •To address the background noise, a spiking autoencoder network is developed using the noise resilience and time-sequence sensitivity of SNNs.•This is the first instance of solving background subtraction from a spike-based perspective, where a continuous spiking convolutional and deconvolutional block is employed to enhance foreground features and diminish background noise within the decoder.•To achieve energy efficiency, a novel self-distillation spiking supervised learning method is proposed within ANN-to-SNN framework.•The empirical evaluations on CDnet-2014 and DAVIS-2016 demonstrate the superiority of the proposed method.
ISSN:0031-3203
DOI:10.1016/j.patcog.2025.111792