Traffic sign recognition method based on lightweight neural network

US11875576B2 · kind B2 · utility

Assignee

• Quanzhou Equipment Manufacturing Research Institute · CN

Inventors

• Jielong Guo · Hangzhou City, CN
• Xian Wei · Nanhu, CN
• Xuan Tang · Hopkinton, US
• Hui Yu · Nanhu, CN
• Jianfeng Zhang · Shrewsbury, US
• Dongheng Shao · Fujian, CN
• Xiaodi Yang · Jinjiang, CN
• Yufang XIE · Jinjiang, CN

Key dates

• Filing date: Jun 23, 2023
• Grant date: Jan 16, 2024
• Expiry date: Jun 23, 2043

Classification

• Technology area (CPC Y): Emerging Cross-Sectional Technologies
• CPC primary: Y02T10/40
• WIPO field: Computer technology
• WIPO sector: Electrical engineering

Abstract

Provided is a traffic sign recognition method based on a lightweight neural network, which including: a lightweight neural network model is constructed for training and pruning to obtain a lightweight neural network model; the lightweight neural network model comprises a convolution feature extraction part and a classifier part; the convolution feature extraction part includes one layer of conventional 3×3 convolution and 16 layers of separable asymmetric convolution. The classifier part includes three layers of separable full connection modules.