Design of Low-Power, High-Precision, and Lightweight Image Recognition System for Multiple Scenes

Aiming at the existing handwritten digit recognition systems with low recognition accuracy, high system power consumption, and high hardware resource consumption, this paper proposes a low-power, high-precision, and lightweight handwritten digit recognition hardware acceleration scheme for multiscen...

Full description

Saved in:
Bibliographic Details
Main Authors: Gangfeng Yang, Liming Chen, Jianhua Jiang, Chengying Chen, Xindong Huang
Format: Article
Language:English
Published: Wiley 2025-01-01
Series:Active and Passive Electronic Components
Online Access:http://dx.doi.org/10.1155/apec/2070758
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849430616662081536
author Gangfeng Yang
Liming Chen
Jianhua Jiang
Chengying Chen
Xindong Huang
author_facet Gangfeng Yang
Liming Chen
Jianhua Jiang
Chengying Chen
Xindong Huang
author_sort Gangfeng Yang
collection DOAJ
description Aiming at the existing handwritten digit recognition systems with low recognition accuracy, high system power consumption, and high hardware resource consumption, this paper proposes a low-power, high-precision, and lightweight handwritten digit recognition hardware acceleration scheme for multiscenario based on FPGA. By optimizing the network structure of a convolutional neural network (CNN) and the number of parameters of the model, this scheme proposes a high-precision and lightweight network model, simplified CNN, and by optimizing the data access mode and memory usage, and by adopting the strategies of time-sharing and multiplexing, weights sharing, and parallel processing for the hardware acceleration of the algorithm, it effectively reduces the consumption of hardware resources and improves the performance of the system. The experimental results show that the recognition accuracy of the algorithm reaches 98.82%, the system response time is 0.481316 ms under the 33 + 200 MHz system clock, and the on-chip power consumption of the system is 0.843 W, and it can be used for real-time handwritten digit recognition in many occasions.
format Article
id doaj-art-646d94e32f0249c8b6f0d2e4e5dbc0f8
institution Kabale University
issn 1563-5031
language English
publishDate 2025-01-01
publisher Wiley
record_format Article
series Active and Passive Electronic Components
spelling doaj-art-646d94e32f0249c8b6f0d2e4e5dbc0f82025-08-20T03:27:56ZengWileyActive and Passive Electronic Components1563-50312025-01-01202510.1155/apec/2070758Design of Low-Power, High-Precision, and Lightweight Image Recognition System for Multiple ScenesGangfeng Yang0Liming Chen1Jianhua Jiang2Chengying Chen3Xindong Huang4School of Opto-Electronic and Communication EngineeringSchool of Integrated CircuitsSchool of ElectronicsSchool of Opto-Electronic and Communication EngineeringSchool of Opto-Electronic and Communication EngineeringAiming at the existing handwritten digit recognition systems with low recognition accuracy, high system power consumption, and high hardware resource consumption, this paper proposes a low-power, high-precision, and lightweight handwritten digit recognition hardware acceleration scheme for multiscenario based on FPGA. By optimizing the network structure of a convolutional neural network (CNN) and the number of parameters of the model, this scheme proposes a high-precision and lightweight network model, simplified CNN, and by optimizing the data access mode and memory usage, and by adopting the strategies of time-sharing and multiplexing, weights sharing, and parallel processing for the hardware acceleration of the algorithm, it effectively reduces the consumption of hardware resources and improves the performance of the system. The experimental results show that the recognition accuracy of the algorithm reaches 98.82%, the system response time is 0.481316 ms under the 33 + 200 MHz system clock, and the on-chip power consumption of the system is 0.843 W, and it can be used for real-time handwritten digit recognition in many occasions.http://dx.doi.org/10.1155/apec/2070758
spellingShingle Gangfeng Yang
Liming Chen
Jianhua Jiang
Chengying Chen
Xindong Huang
Design of Low-Power, High-Precision, and Lightweight Image Recognition System for Multiple Scenes
Active and Passive Electronic Components
title Design of Low-Power, High-Precision, and Lightweight Image Recognition System for Multiple Scenes
title_full Design of Low-Power, High-Precision, and Lightweight Image Recognition System for Multiple Scenes
title_fullStr Design of Low-Power, High-Precision, and Lightweight Image Recognition System for Multiple Scenes
title_full_unstemmed Design of Low-Power, High-Precision, and Lightweight Image Recognition System for Multiple Scenes
title_short Design of Low-Power, High-Precision, and Lightweight Image Recognition System for Multiple Scenes
title_sort design of low power high precision and lightweight image recognition system for multiple scenes
url http://dx.doi.org/10.1155/apec/2070758
work_keys_str_mv AT gangfengyang designoflowpowerhighprecisionandlightweightimagerecognitionsystemformultiplescenes
AT limingchen designoflowpowerhighprecisionandlightweightimagerecognitionsystemformultiplescenes
AT jianhuajiang designoflowpowerhighprecisionandlightweightimagerecognitionsystemformultiplescenes
AT chengyingchen designoflowpowerhighprecisionandlightweightimagerecognitionsystemformultiplescenes
AT xindonghuang designoflowpowerhighprecisionandlightweightimagerecognitionsystemformultiplescenes