A Hardware-Oriented Algorithm for Ultra-High-Speed Object Detection

doi:10.1109/JSEN.2019.2895294

	A Hardware-Oriented Algorithm for Ultra-High-Speed Object Detection
	Li, Jianquan1,2 ; Liu, Xilong1,2 ; Liu, Fangfang1,2 ; Xu, De1,2 ; Gu, Qingyi 1,2; Ishii, Idaku 3
刊名	IEEE SENSORS JOURNAL
	2019-05-15
卷号	19 期号:10 页码:3818-3831
关键词	Hardware implementation high-frame-rate vision field-programmable gate array multi-object detection histograms of oriented gradient support vector machine
ISSN号	1530-437X
DOI	10.1109/JSEN.2019.2895294
通讯作者	Gu, Qingyi(qingyi.gu@ia.ac.cn)
英文摘要	This paper describes a novel hardware-oriented algorithm that can be implemented on a field-programmable gate array in a high-speed vision platform for detection of multiple objects with clear texture information in images of 512 x 512 pixels at 10000 frames per second (fps) under complex background. The proposed algorithm is specially designed for devices with limited hardware resource for high-frame-rate, high-data-throughput, and high-parallelism processing of video streams with low latency. The proposed algorithm is based on the conventional histograms of oriented gradient (HOC) descriptor and support vector machine classifier algorithms. Considering the trade-off between speed and accuracy, many hardware-based optimization operations were implemented. The data throughput is nearly 29.30 Gbps while the latency for feature extraction is 0.76 us (61 clock period). After hardware-based image processing, the source image and the detected object features can be transferred to a personal computer for recording or post-processing at 10000 fps. Several experiments were done to demonstrate the performance of our proposed algorithm for ultra-high-speed moving object detection with clear texture information in images.
资助项目	National Natural Science Foundation of China[61673376] ; National Natural Science Foundation of China[61503376] ; Emergency Management Project of the National Natural Science Foundation of China[61741317]
WOS关键词	FEATURES
WOS研究方向	Engineering ; Instruments & Instrumentation ; Physics
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:000465288100029
资助机构	National Natural Science Foundation of China ; Emergency Management Project of the National Natural Science Foundation of China
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/24922]
专题	中国科学院自动化研究所
通讯作者	Gu, Qingyi
作者单位	1.Chinese Acad Sci, Res Ctr Precis Sensing & Control, Inst Automat, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Sch Comp & Control Engn, Beijing 101408, Peoples R China 3.Hiroshima Univ, Dept Syst Cybernet, Robot Lab, Hiroshima 7398527, Japan
推荐引用方式 GB/T 7714	Li, Jianquan,Liu, Xilong,Liu, Fangfang,et al. A Hardware-Oriented Algorithm for Ultra-High-Speed Object Detection[J]. IEEE SENSORS JOURNAL,2019,19(10):3818-3831.
APA	Li, Jianquan,Liu, Xilong,Liu, Fangfang,Xu, De,Gu, Qingyi,&Ishii, Idaku.(2019).A Hardware-Oriented Algorithm for Ultra-High-Speed Object Detection.IEEE SENSORS JOURNAL,19(10),3818-3831.
MLA	Li, Jianquan,et al."A Hardware-Oriented Algorithm for Ultra-High-Speed Object Detection".IEEE SENSORS JOURNAL 19.10(2019):3818-3831.