Hybrid Optimization of Target Detection on Embedded Platforms for Real Time Applications

Xinchen Zhang; Wangchao Sun; Yaodong Zhao; Kaisheng Liao; Yilin Liu; Hongda Xu; Zhuoling Xiao; Bo Yan

doi:10.1109/ICET55676.2022.9825328

Hybrid Optimization of Target Detection on Embedded Platforms for Real Time Applications

Xinchen Zhang, Wangchao Sun, Yaodong Zhao, Kaisheng Liao, Yilin Liu, Hongda Xu, Zhuoling Xiao^*, Bo Yan

^*Corresponding author for this work

University of Electronic Science and Technology of China

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Target detection has been widely used in fields such as intelligent security and autonomous driving. However, existing computationally heavy target detection algorithms based on deep learning can only work on GPU and CPU platforms, restricting the applications on edge devices with limited computational power. To address this issue, this paper proposes layer fusion and 16-bit fixed-point quantization on the YOLOv2-Tiny algorithm to reduce the computational complexity of target detection algorithms. Furthermore, the data transmission efficiency is optimized by using ping-pong butter and multi-channel methods. To reduce FPGA resource consumption, the neural network is split into convolution, accumulation, pooling, and address mapping modules. The proposed system has been successfully implemented on the Xilinx Zynq-XC7Z035 platform, using only 47% of BRAM resources and 18% of DSP resources.

Original language	English
Title of host publication	2022 IEEE 5th International Conference on Electronics Technology, ICET 2022
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1136-1141
Number of pages	6
ISBN (Electronic)	9781665485081
DOIs	https://doi.org/10.1109/ICET55676.2022.9825328
Publication status	Published - 2022
Externally published	Yes
Event	5th IEEE International Conference on Electronics Technology, ICET 2022 - Chengdu, China Duration: 13 May 2022 → 16 May 2022

Publication series

Name	2022 IEEE 5th International Conference on Electronics Technology, ICET 2022

Conference

Conference	5th IEEE International Conference on Electronics Technology, ICET 2022
Country/Territory	China
City	Chengdu
Period	13/05/22 → 16/05/22

Keywords

target detection
YOLOv2-Tiny
Zynq

Access to Document

10.1109/ICET55676.2022.9825328

Cite this

Zhang, X., Sun, W., Zhao, Y., Liao, K., Liu, Y., Xu, H., Xiao, Z., & Yan, B. (2022). Hybrid Optimization of Target Detection on Embedded Platforms for Real Time Applications. In 2022 IEEE 5th International Conference on Electronics Technology, ICET 2022 (pp. 1136-1141). (2022 IEEE 5th International Conference on Electronics Technology, ICET 2022). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICET55676.2022.9825328

Zhang, Xinchen ; Sun, Wangchao ; Zhao, Yaodong et al. / Hybrid Optimization of Target Detection on Embedded Platforms for Real Time Applications. 2022 IEEE 5th International Conference on Electronics Technology, ICET 2022. Institute of Electrical and Electronics Engineers Inc., 2022. pp. 1136-1141 (2022 IEEE 5th International Conference on Electronics Technology, ICET 2022).

@inproceedings{e5dcebb618d043d591e13a44a6fa6c41,

title = "Hybrid Optimization of Target Detection on Embedded Platforms for Real Time Applications",

abstract = "Target detection has been widely used in fields such as intelligent security and autonomous driving. However, existing computationally heavy target detection algorithms based on deep learning can only work on GPU and CPU platforms, restricting the applications on edge devices with limited computational power. To address this issue, this paper proposes layer fusion and 16-bit fixed-point quantization on the YOLOv2-Tiny algorithm to reduce the computational complexity of target detection algorithms. Furthermore, the data transmission efficiency is optimized by using ping-pong butter and multi-channel methods. To reduce FPGA resource consumption, the neural network is split into convolution, accumulation, pooling, and address mapping modules. The proposed system has been successfully implemented on the Xilinx Zynq-XC7Z035 platform, using only 47% of BRAM resources and 18% of DSP resources.",

keywords = "target detection, YOLOv2-Tiny, Zynq",

author = "Xinchen Zhang and Wangchao Sun and Yaodong Zhao and Kaisheng Liao and Yilin Liu and Hongda Xu and Zhuoling Xiao and Bo Yan",

note = "Publisher Copyright: {\textcopyright} 2022 IEEE.; 5th IEEE International Conference on Electronics Technology, ICET 2022 ; Conference date: 13-05-2022 Through 16-05-2022",

year = "2022",

doi = "10.1109/ICET55676.2022.9825328",

language = "English",

series = "2022 IEEE 5th International Conference on Electronics Technology, ICET 2022",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "1136--1141",

booktitle = "2022 IEEE 5th International Conference on Electronics Technology, ICET 2022",

address = "United States",

}

Zhang, X, Sun, W, Zhao, Y, Liao, K, Liu, Y, Xu, H, Xiao, Z & Yan, B 2022, Hybrid Optimization of Target Detection on Embedded Platforms for Real Time Applications. in 2022 IEEE 5th International Conference on Electronics Technology, ICET 2022. 2022 IEEE 5th International Conference on Electronics Technology, ICET 2022, Institute of Electrical and Electronics Engineers Inc., pp. 1136-1141, 5th IEEE International Conference on Electronics Technology, ICET 2022, Chengdu, China, 13/05/22. https://doi.org/10.1109/ICET55676.2022.9825328

Hybrid Optimization of Target Detection on Embedded Platforms for Real Time Applications. / Zhang, Xinchen; Sun, Wangchao; Zhao, Yaodong et al.
2022 IEEE 5th International Conference on Electronics Technology, ICET 2022. Institute of Electrical and Electronics Engineers Inc., 2022. p. 1136-1141 (2022 IEEE 5th International Conference on Electronics Technology, ICET 2022).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Hybrid Optimization of Target Detection on Embedded Platforms for Real Time Applications

AU - Zhang, Xinchen

AU - Sun, Wangchao

AU - Zhao, Yaodong

AU - Liao, Kaisheng

AU - Liu, Yilin

AU - Xu, Hongda

AU - Xiao, Zhuoling

AU - Yan, Bo

PY - 2022

Y1 - 2022

N2 - Target detection has been widely used in fields such as intelligent security and autonomous driving. However, existing computationally heavy target detection algorithms based on deep learning can only work on GPU and CPU platforms, restricting the applications on edge devices with limited computational power. To address this issue, this paper proposes layer fusion and 16-bit fixed-point quantization on the YOLOv2-Tiny algorithm to reduce the computational complexity of target detection algorithms. Furthermore, the data transmission efficiency is optimized by using ping-pong butter and multi-channel methods. To reduce FPGA resource consumption, the neural network is split into convolution, accumulation, pooling, and address mapping modules. The proposed system has been successfully implemented on the Xilinx Zynq-XC7Z035 platform, using only 47% of BRAM resources and 18% of DSP resources.

AB - Target detection has been widely used in fields such as intelligent security and autonomous driving. However, existing computationally heavy target detection algorithms based on deep learning can only work on GPU and CPU platforms, restricting the applications on edge devices with limited computational power. To address this issue, this paper proposes layer fusion and 16-bit fixed-point quantization on the YOLOv2-Tiny algorithm to reduce the computational complexity of target detection algorithms. Furthermore, the data transmission efficiency is optimized by using ping-pong butter and multi-channel methods. To reduce FPGA resource consumption, the neural network is split into convolution, accumulation, pooling, and address mapping modules. The proposed system has been successfully implemented on the Xilinx Zynq-XC7Z035 platform, using only 47% of BRAM resources and 18% of DSP resources.

KW - target detection

KW - YOLOv2-Tiny

KW - Zynq

UR - http://www.scopus.com/inward/record.url?scp=85136329765&partnerID=8YFLogxK

U2 - 10.1109/ICET55676.2022.9825328

DO - 10.1109/ICET55676.2022.9825328

M3 - Conference contribution

AN - SCOPUS:85136329765

T3 - 2022 IEEE 5th International Conference on Electronics Technology, ICET 2022

SP - 1136

EP - 1141

BT - 2022 IEEE 5th International Conference on Electronics Technology, ICET 2022

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 5th IEEE International Conference on Electronics Technology, ICET 2022

Y2 - 13 May 2022 through 16 May 2022

ER -

Zhang X, Sun W, Zhao Y, Liao K, Liu Y, Xu H et al. Hybrid Optimization of Target Detection on Embedded Platforms for Real Time Applications. In 2022 IEEE 5th International Conference on Electronics Technology, ICET 2022. Institute of Electrical and Electronics Engineers Inc. 2022. p. 1136-1141. (2022 IEEE 5th International Conference on Electronics Technology, ICET 2022). doi: 10.1109/ICET55676.2022.9825328

Hybrid Optimization of Target Detection on Embedded Platforms for Real Time Applications

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this