Improved SSD Algorithm and Its Performance Analysis of Small Target Detection in Remote Sensing Images

Fig. 3. Comparison of feature maps before and after integration. (a) Input image; (b) output of dense block2; (c) output of dense block3; (d) output of dense block2 with feature integration; (e) output of dense block3 with feature integration; (f) output of dense block4

Download full size

Fig. 4. Interface of training sample online acquisition system. (a) Superimposed main airport point data; (b) aircraft sample collection

Download full size

Fig. 5. Size of each target in sample set

Download full size

Fig. 6. Decay curve of learning rate

Download full size

Fig. 7. Comparison of total loss and precision between transfer training and random initialization. (a) Total loss varies with number of iterations; (b) MAPRIoU=0.50 varies with number of iterations

Download full size

Fig. 8. Comparison of precisions of improved SSD algorithm and other algorithms varying with number of iterations. (a) MAP; (b) MAPlarge; (c) MAPmedium; (d) MAPsmall; (e) MAPRIoU=0.50; (f) MAPRIoU=0.75

Download full size

Fig. 9. Comparison of improved SSD algorithm and other algorithms in detection effect. (a) Faster R-CNN+ResNet101; (b) R-FCN+ResNet101; (c) improved SSD algorithm

Download full size

Table 1. Main metrics

View table

Table 1. Main metrics

Metric	Remarks
MAP	MAP at R_IoU in {0.5+0.05×m,m=0,1,…,9} (primary challenge metric)
$MA P^{R_{IoU} = 0.50}$	MAP at R_IoU=0.50 (pascal VOC metric)
$MA P^{R_{IoU} = 0.75}$	MAP at R_IoU=0.75 (strict metric)
MAP^small	MAP for small targets: S_area<(32 pixel)²
MAP^medium	MAP for medium targets: (32 pixel)²≤S_area≤(96 pixel)²
MAP^large	MAP for large targets: S_area>(96 pixel)²

Table 2. Sample set statistics

View table

Table 2. Sample set statistics

Data set	Class	Target amount					Percentage /%
Data set	Class	Small	Medium	Large	Total		Small		Medium		Large
Training set	airplane	1204	1542	431	4401	27.36		35.04		9.79
Training set	playground	178	516	530	4401	4.04		11.72		12.04
Validation set	airplane	390	427	74	1169	33.36		36.53		6.33
Validation set	playground	27	120	131	1169	2.31		10.27		14.21
Test set	airplane	940	366	111	1892	49.68		19.34		5.87
Test set	playground	133	294	48	1892	7.03		15.54		2.54

Table 3. Comparison of calculation time and precision on validation set

View table

Table 3. Comparison of calculation time and precision on validation set

Method	Parameter /MB	Time overhead /ms	Metric /%
Method	Parameter /MB	Time overhead /ms	MAP	MAP^large	MAP^medium	MAP^small	$MA P^{R_{IoU} = 0.50}$	$MA P^{R_{IoU} = 0.75}$
SSD+Inceptionv2	53.4	24.8	47.15	69.48	50.16	11.56	85.08	48.95
Faster R-CNN+ResNet50	173.3	108.6	47.12	72.81	48.91	10.16	82.72	50.59
Faster R-CNN+ResNet101	249.5	117.5	50.50	73.06	53.51	13.76	85.84	55.03
R-FCN+ResNet101	258.2	79.3	51.17	74.69	51.43	16.01	87.20	55.86
Improved SSD algorithm	59.8	71.8	54.14	73.31	54.32	21.16	90.55	57.38

Table 4. Comparison of precision on test set

View table

Table 4. Comparison of precision on test set

Methods	Metric /%
Methods	MAP	MAP^large	MAP^medium	MAP^small	$MA P^{R_{IoU} = 0.50}$	$MA P^{R_{IoU} = 0.75}$
SSD+Inceptionv2	35.85	62.73	43.55	19.52	77.72	24.07
Faster R-CNN+ResNet50	28.72	61.53	34.67	13.43	68.87	17.60
Faster R-CNN+ResNet101	36.05	61.55	42.83	21.74	76.83	27.69
R-FCN+ResNet101	36.70	59.18	43.96	22.91	77.30	28.23
Improved SSD algorithm	45.18	65.31	50.68	31.65	83.95	42.15

Tools

Get Citation

Copy Citation Text

Junqiang Wang, Jiansheng Li, Xuewen Zhou, Xu Zhang. Improved SSD Algorithm and Its Performance Analysis of Small Target Detection in Remote Sensing Images[J]. Acta Optica Sinica, 2019, 39(6): 0628005

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites

Paper Information

Category: Remote Sensing and Sensors

Received: Jan. 16, 2019

Accepted: Mar. 12, 2019

Published Online: Jun. 17, 2019

The Author Email: Li Jiansheng (xindawangjunqiang@163.com)

DOI:10.3788/AOS201939.0628005

Topics

laser devices and laser physics

Lasers and Laser Optics

Laser physics

laser manufacturing

Instrumentation, Measurement and Metrology