Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
CONVOLUTIONAL NEURAL NETWORK DEPLOYMENT AND OPTIMIZATION METHOD FOR MICROCONTROLLER
Document Type and Number:
WIPO Patent Application WO/2023/236319
Kind Code:
A1
Abstract:
A convolutional neural network deployment and optimization method for a microcontroller, comprising three parts: design of a convolutional neural network model, optimization of a convolutional calculation memory, and deployment of a convolutional neural network. The design of the convolutional neural network model is based on neural network architecture search, so as to search for a convolutional neural network model that is suitable for a microcontroller and has small calculation amount, parameter quantity, and memory requirements; standard convolution, deep convolution, and point convolution that are commonly used in a convolutional neural network are respectively optimized to reduce the memory occupation in a convolutional neural network reasoning process, so that the convolutional neural network can run on multiple memory-constrained microcontrollers; and a method for construction and application of a convolutional neural network running on a microcontroller is provided, so that the ease to use and practicality of the microcontroller running a convolutional neural network model are improved.

More Like This:
WO/2022/156475NEURAL NETWORK MODEL TRAINING METHOD AND APPARATUS, AND DATA PROCESSING METHOD AND APPARATUS
WO/2017/178666AUTONOMOUS SET OF DEVICES AND METHOD FOR DETECTING AND IDENTIFYING PLANT SPECIES IN AN AGRICULTURAL CROP FOR THE SELECTIVE APPLICATION OF AGROCHEMICALS
WO/2023/171886DEEP LEARNING-BASED MOLECULE DESIGN METHOD, AND DEVICE AND COMPUTER PROGRAM FOR PERFORMING SAME
Inventors:
SUN YANFEI (CN)
WANG ZINIU (CN)
QI JIN (CN)
Application Number:
PCT/CN2022/106634
Publication Date:
December 14, 2023
Filing Date:
July 20, 2022
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
UNIV NANJING POSTS & TELECOMMUNICATIONS (CN)
International Classes:
G06N3/04
Foreign References:
CN114742211A2022-07-12
CN109447239A2019-03-08
CN112766467A2021-05-07
US20190188537A12019-06-20
Other References:
ZHANG, YINGJIE: "Design and Implementation of Deep Convolutional Neural Network Embedded Inference Framework", CHINESE MASTER’S THESES FULL-TEXT DATABASE, 15 February 2021 (2021-02-15)
Attorney, Agent or Firm:
NANJING ZHENGLIAN INTELLECTUAL PROPERTY AGENCY CO., LTD. (CN)
Download PDF:
View/Download PDF      PDF Help



 
Previous Patent: FLUE GAS TREATMENT PROCESS

Next Patent: FOLDING STRUCTURE AND FOLDING TELESCOPIC FRAME