Object Detection in Optical Remote Sensing Images Based on FFC-SSD Model

Junda Xue; Jiajia Zhu; Jing Zhang; Xiaohui Li; Shuai Dou; Lin Mi; Ziyang Li; Xinfang Yuan; Chuanrong Li

doi:10.3788/AOS202242.1210002

Acta Optica Sinica, Volume. 42, Issue 12, 1210002(2022)

Object Detection in Optical Remote Sensing Images Based on FFC-SSD Model

Junda Xue^1,2, Jiajia Zhu^1,2、**, Jing Zhang^1,2、*, Xiaohui Li^1、***, Shuai Dou¹, Lin Mi¹, Ziyang Li¹, Xinfang Yuan¹, and Chuanrong Li¹

Author Affiliations

¹Aerospace Information Research Institute, Key Laboratory of Quantitative Remote Sensing Information Technology, Chinese Academy of Sciences, Beijing 100094, China

²University of Chinese Academy of Sciences, Beijing 100049, China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(16)

Fig. 1. Framework of FFC-SSD model

Download full size

Fig. 2. Number of samples and size distribution of each category in DOTA dataset. (a) Number of samples; (b) size distribution

Download full size

Fig. 3. Average coverage of five groups varying with number of clusters k

Download full size

Fig. 4. Distribution of sample target box dimensions in DOTA and default target box dimensions set by group clustering

Download full size

Fig. 5. Diagram of MSFF module

Download full size

Fig. 6. Output feature maps of MSFF_D and MSFF_U modules. (a) Original images; (b) output feature maps of MSFF_D module; (c) output feature maps of MSFF_U module

Download full size

Fig. 7. Convergence curves of loss function

Download full size

Fig. 8. Average precision (AP) for each category in DOTA testing dataset for each experiment

Download full size

Fig. 9. Test results of SSD and FFC-SSD models. (a)(c) SSD; (b)(d) FFC-SSD

Download full size

Table 1. Grouping description of target categories in DOTA datasets

View table

Table 1. Grouping description of target categories in DOTA datasets

Group	Object category	Sample number per category
T0	Small vehicle (SV)	>100000
T1	Large vehicle (LV), ship	20000~40000
T2	Plane, storage tank (ST), harbor	5000~10000
T3	Bridge, tennis court (TC), swimming pool (SP)	2000~5000
T4	Roundabout (RA), soccer field(SF), ground field track (GFT), Baseball diamond (BD), basketball court (BC), helicopter (HC)	<2000

Table 2. Default target box size on each fusion feature map

View table

Table 2. Default target box size on each fusion feature map

n	Layer	Feature map size /(pixel×pixel)	Size of default box w×h /(pixel×pixel)
1	Conv4_3	256×256	5×10,10×6,12×21,20×11,14×13,22×20,18×27,36×17
2	Conv5_3	128×128	27×39,75×45,48×29,42×69,40×39,84×23,24×66
3	FC7	64×64	72×80,92×89,56×72,48×91,105×60,73×100
4	Conv8_2	32×32	149×96,159×152,40×133,130×120
5	Conv9_2	16×16	167×201,97×187,59×210
6	Conv10_2	8×8	246×248
7	Conv11_2	4×4	290×323

Table 3. Comparsion of average coverage for each category in DOTA of default object frame parameters set by two methods%

View table

Table 3. Comparsion of average coverage for each category in DOTA of default object frame parameters set by two methods%

Object category	SV	LV	Ship	Plane	ST	Harbor	Bridge	TC
SSD	35.65	66.11	66.29	80.12	54.04	76.20	59.25	77.27
BGC	80.99	78.09	80.10	87.24	87.31	78.77	80.21	86.94
Object category	RA	SF	HP	GFT	BD	SP	BC
SSD	77.36	67.46	66.84	78.96	85.21	73.63	77.70
BGC	88.87	79.46	79.20	80.51	88.06	84.56	82.48

Table 4. Influence of each module on mAP and FPS of object detection
View table
Table 4. Influence of each module on mAP and FPS of object detection
Experiment No. Model Aps /% APless /% mAP /% FPS
1 SSD 33.5 52.5 55.9 26
2 SSD+MSFF_U 44.3 63.5 64.6 24
3 SSD+BGC 49.5 62.4 63.6 16
4 SSD+BGC+MSFF_U (FFC-SSD) 69.3 69.9 74.9 15
5 SSD+BGC+MSFF_D 63.4 64.4 70.0 12

Table 5. Detection performance of FFC-SSD and other models on DOTA dataset%

View table

Table 5. Detection performance of FFC-SSD and other models on DOTA dataset%

Model		SSD^[1]	YOLOv3^[9]	FRCNN^[2,20]	DSSD^[15,21]	FMSSD^[15]	FFC-SSD
	Plane	84.2	91.0	80.3	91.1	89.1	88.4
	Small vehicle	39.9	40.3	53.6	79.0	69.2	82.5
	Large vehicle	55.9	76.9	52.5	77.2	73.6	76.4
	Roundabout	52.6	58.5	49.8	72.6	67.5	74.1
	Bridge	25.7	50.0	32.9	54.6	48.2	51.0
	Soccer field	56.7	18.0	57.0	38.0	52.7	62.0
	Helicopter	33.0	85.2	41.9	28.9	60.2	54.3
AP	Ground field track	54.8	30.2	68.1	66.4	68.0	74.7
	Baseball diamond	72.7	68.3	77.6	71.8	81.5	78.3
	Storage tank	61.7	82.1	59.6	69.7	73.3	87.2
	Tennis court	80.4	92.0	90.4	87.6	90.7	90.6
	Swimming pool	62.0	80.2	56.5	59.4	80.6	73.0
	Ship	65.9	89.2	50.0	87.5	76.9	87.4
	Harbor	48.4	69.3	61.7	75.4	72.4	67.2
	Basketball court	45.3	62.4	75.1	52.1	82.7	76.2
mAP		55.9	66.2	60.6	67.4	72.4	74.9
stdAP		15.8	22.2	14.9	17.4	11.7	11.6
FPS		26	13	7	9	16	15

Table 6. Detection performance of FFC-SSD and other models on NWPU VHR-10 dataset%

View table

Table 6. Detection performance of FFC-SSD and other models on NWPU VHR-10 dataset%

Model		SSD^[1]	YOLOv3^[9]	FRCNN^[2,15]	FMSSD^[15]	FFC-SSD
	Plane	98.2	95.6	94.6	99.7	99.7
	Ship	83.9	88.6	82.3	89.9	96.3
	Storage tank	75.9	77.9	65.3	90.3	88.1
	Baseball diamond	90.2	91.7	95.5	98.2	99.4
AP	Tennis court	85.6	89.1	81.9	86.0	90.3
	Basketball court	79.6	89.8	89.7	96.8	99.4
	Ground track field	92.2	84.8	92.4	99.6	99.9
	Harbor	77.1	81.2	72.4	75.6	96.1
	Bridge	67.8	70.8	57.5	80.1	98.5
	Vehicle	75.6	87.8	77.8	88.2	89.0
mAP		82.6	85.7	80.9	90.4	95.7
stdAP		8.70	6.92	12.19	7.89	4.49

Table 7. mAP of different models on three optical remote sensing datasets%
View table
Table 7. mAP of different models on three optical remote sensing datasets%
Model NWPU VHR-10 RSOD UCAS-AOD
SSD 61.7 45.1 48.9
FFC-SSD 76.5 60.2 69.6

Tools

Get Citation

Copy Citation Text

Junda Xue, Jiajia Zhu, Jing Zhang, Xiaohui Li, Shuai Dou, Lin Mi, Ziyang Li, Xinfang Yuan, Chuanrong Li. Object Detection in Optical Remote Sensing Images Based on FFC-SSD Model[J]. Acta Optica Sinica, 2022, 42(12): 1210002

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites

Paper Information

Category: Image Processing

Received: Sep. 24, 2021

Accepted: Nov. 25, 2021

Published Online: Jun. 20, 2022

The Author Email: Zhu Jiajia (jjzhu@aoe.ac.cn), Zhang Jing (zhangjing@aoe.ac.cn), Li Xiaohui (xhli@aoe.ac.cn)

DOI:10.3788/AOS202242.1210002

Topics

laser devices and laser physics

Lasers and Laser Optics

Laser physics

laser manufacturing

Instrumentation, Measurement and Metrology

Table 1. Grouping description of target categories in DOTA datasets

Table 1. Grouping description of target categories in DOTA datasets

Table 2. Default target box size on each fusion feature map

Table 2. Default target box size on each fusion feature map

Table 3. Comparsion of average coverage for each category in DOTA of default object frame parameters set by two methods%

Table 3. Comparsion of average coverage for each category in DOTA of default object frame parameters set by two methods%

Table 4. Influence of each module on mAP and FPS of object detection

Table 4. Influence of each module on mAP and FPS of object detection

Table 5. Detection performance of FFC-SSD and other models on DOTA dataset%

Table 5. Detection performance of FFC-SSD and other models on DOTA dataset%

Table 6. Detection performance of FFC-SSD and other models on NWPU VHR-10 dataset%

Table 6. Detection performance of FFC-SSD and other models on NWPU VHR-10 dataset%

Table 7. mAP of different models on three optical remote sensing datasets%

Table 7. mAP of different models on three optical remote sensing datasets%