Low-Altitude UAV Detection and Recognition Method Based on Optimized YOLOv3

Qi Ma; Bin Zhu; Hongwei Zhang; Yang Zhang; Yuchen Jiang

doi:10.3788/LOP56.201006

Laser & Optoelectronics Progress, Volume. 56, Issue 20, 201006(2019)

Low-Altitude UAV Detection and Recognition Method Based on Optimized YOLOv3

Qi Ma^1,2、*, Bin Zhu^1,2、**, Hongwei Zhang^1,2、***, Yang Zhang^1,2, and Yuchen Jiang^1,2

¹State Key Laboratory of Pulsed Power Laser Technology, College of Electronic Engineering, National University of Defense Technology, Hefei, Anhui 230037, China

²National University of Defense Technology, Hefei, Anhui 230037, China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(7)

Fig. 1. Structural diagram of detection and recognition for low-altitude UAV

Download full size

Fig. 2. Residual block structural diagram

Download full size

Fig. 3. Diagrams of 12 anchors with 4 scales. At each scale, the solid box is a corresponding square, and the dotted boxes are the corresponding anchor boxes. (a) 10×10; (b) 20×20; (c) 40×40; (d) 80×80

Download full size

Fig. 4. Comparison of partial detection resultson test dataset (The first and third columns are the results of the YOLOv3 method, and the second and fourth columns are the results of the O-YOLOv3 method). (a) DJ-Air; (b) DJ-Pro; (c) DJ-3

Download full size

Table 1. O-YOLO network structure for detection and recognition of low altitude UAV

View table

Table 1. O-YOLO network structure for detection and recognition of low altitude UAV

Repetition times	Type	Filters	Size	Output
	Convolutional	16	3×3	640×640×16
	Convolutional	32	3×3/2	320×320×32
	Convolutional	16	1×1
1×	Convolutional	32	3×3
	Residual			320×320×32
	Convolutional	64	3×3/2	160×160×64
	Convolutional	32	1×1
2×	Convolutional	64	3×3
	Residual			160×160×64
1×	Convolutional	128	3×3/2	80×80×128
	Convolutional	64	1×1
8×	Convolutional	128	3×3
	Residual			80×80×128
	Convolutional	256	3×3/2	40×40×256
	Convolutional	128	1×1
8×	Convolutional	256	3×3
	Residual			40×40×256
1×	Convolutional	512	3×3/2	20×20×512
	Convolutional	256	1×1
4×	Convolutional	512	3×3
	Residual			20×20×512
1×	Convolutional	1024	3×3/2	10×10×1024
	Convolutional	512	1×1
4×	Convolutional	1024	3×3
	Residual			10×10×1024
	Avgpool		Global
1×	Connected		3
	Softmax

Table 2. Processing methods and sample numbers for different train datasets
View table
Table 2. Processing methods and sample numbers for different train datasets
Traindataset Processing method Numberof samples
UAV_A Original 2400
UAV_B Original + Image enhancement 4800
UAV_C Original + Image enhancement +Data augmentation 46000

Table 3. Evaluation results of different methods on different test datasets

View table

Table 3. Evaluation results of different methods on different test datasets

Train set	Method	Detection speed /s^-1	mAP /%	AP /%
Train set	Method	Detection speed /s^-1	mAP /%	DJ-Air	DJ-Pro	DJ-3
UAV_A	YOLOv3	30.28	70.11	64.14	60.62	85.57
	O-YOLOv3	26.13	77.87	73.86	69.12	90.63
UAV_B	YOLOv3	29.87	71.24	64.56	62.13	87.02
	O-YOLOv3	25.75	79.43	75.01	71.24	92.04
UAV_C	YOLOv3	30.05	73.86	67.23	65.02	89.34
	O-YOLOv3	25.85	82.15	77.86	73.25	95.33

Tools

Get Citation

Copy Citation Text

Qi Ma, Bin Zhu, Hongwei Zhang, Yang Zhang, Yuchen Jiang. Low-Altitude UAV Detection and Recognition Method Based on Optimized YOLOv3[J]. Laser & Optoelectronics Progress, 2019, 56(20): 201006

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites