SinoTechIntel Academic Portal
Official PDF TranslationFrontiers of Information Technology & Electronic Engineering

S3Det: a fast object detector for remote sensing images based on artificial to spiking neural network conversion

Authors: Li Chen; Fan Zhang; Guangwei Xie; Yanzhao Gao; Xiaofeng Qi; Mingqian Sun

DOI: 10.1631/FITEE_2400594Status: Verified Translated Edition
Sponsored AdvertisementAd Placement Area
reCAPTCHA Bot Shield Active

Preparing Secure Academic Download

Verifying human reader & generating high-resolution document...

Verifying Document Integrity15s remaining
← Back to Article
Protected by Google reCAPTCHA v3.PrivacyTerms
Sponsored ContentAdSense In-Feed Ad Slot

Key Findings in This Report

• Proposes S3Det, a novel fast object detector for remote sensing images using ANN-to-SNN conversion. • Introduces a fast sparse model for pulse sequence perception and channel self-decaying weighted normalization (CSWN) to reduce conversion error. • Achieves accuracy comparable to the original ANN while consuming only 1.46 W, reducing power consumption by a factor of 122. • Achieves 24.32% sparsity relative to the benchmark, enabling energy-efficient real-time remote sensing detection.