.png){kind=icon}
.png){kind=icon}
Return
N+Half BNN: An ultra-lightweight binary neural network for traffic signal recognition
.png){kind=icon}
OA
AI
M
Y
J
X
DOI：10.1016/j.icte.2026.05.016.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• We propose an ultra-lightweight BNN for traffic signal recognition with a hardware-oriented N+Half design. • Operation fusion removes most inference-time floating-point operations and reduces storage pressure. • The model reaches 97.64% accuracy on GTSRB while using about 10% of the storage of a full-precision baseline.
Keywords:
Computer vision
Binary neural network
Image classification
Multimedia
Smart cockpit
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
4.2
Papers:
960
Citations:
2.5K
Organization
No organization information available
Cited Papers