Deep Learning and Spatial Analysis Based Port Detection

Zeming Li; Liang Cheng; Daming Zhu; Zhaojin Yan; Chen Ji; Zhixin Duan; Min Jing; Ning Li; Shengkun Dongye; Yanruo Song; Jiahui Liu

doi:10.3788/LOP202158.2028002

Laser & Optoelectronics Progress, Volume. 58, Issue 20, 2028002(2021)

Deep Learning and Spatial Analysis Based Port Detection

Zeming Li¹, Liang Cheng^2,3,4,5, Daming Zhu^1、*, Zhaojin Yan^2,3, Chen Ji^2,3, Zhixin Duan^2,3, Min Jing², Ning Li², Shengkun Dongye¹, Yanruo Song¹, and Jiahui Liu⁶

Author Affiliations

¹Faculty of Land and Resources Engineering, Kunming University of Science and Technology, Kunming, Yunnan 650093, China

²School of Geography and Ocean Science, Nanjing University, Nanjing, Jiangsu 210023, China

³Collaborative Innovation Center of South China Sea Studies, Nanjing, Jiangsu 210023, China

⁴Jiangsu Center for Collaborative Innovation in Geographical Information Resource Development and Application, Nanjing, Jiangsu 210023, China

⁵Jiangsu Center for Collaborative Innovation in Novel Software Technology and Industrialization, Nanjing, Jiangsu 210023, China

⁶School of Geography and Ecotourism, Southwest Forestry University, Kunming, Yunnan 650051, China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(8)

Fig. 1. Structural diagram of YOLO v3 network

Download full size

Fig. 2. Wharf marking at different datasets. (a) DIOR; (b) TGRS-HRRSD; (c)(d) level 19 Google remote sensing image

Download full size

Fig. 3. PR curve of wharf recognition

Download full size

Fig. 4. Recognition results of wharves. (a) Original image; (b) local image; (c) recognition of wharf with single ship docked; (d) recognition of wharf with many ships docked; (e) wharf recognition in complex scene; (f) wharf recognition when prescene of flares on sea surface; (g) recognition of jetty wharf; (h) recognition of along-shore wharf; (i)(j) typical misrecognitions

Download full size

Fig. 5. (a) Port hotspots and (b)--(e) aggregated polygons

Download full size

Table 1. Comparsion of anchor boxes before and after adjustment

View table

Table 1. Comparsion of anchor boxes before and after adjustment

Scale of feature map	Original anchor boxes	New anchor boxes
52×52	10×13; 16×30; 33×23	31×97; 31×30; 47×156
26×26	30×61; 62×45; 59×119	57×49; 85×93; 130×43
13×13	116×90; 156×198; 373×326	164×162; 293×300; 511×511

Table 2. Comparison of performances between two models in experimental area
View table
Table 2. Comparison of performances between two models in experimental area
Model A_TP A_FP A_FN p r F₁
Original YOLO v3 125 87 563 58.96% 18.17% 27.78%
Proposed algorithm 364 137 324 72.65% 52.91% 61.23%

Table 3. Hotspot analysis results under different thresholds

View table

Table 3. Hotspot analysis results under different thresholds

Threshold /m	Number of hotspots	Number of ports	Proportion of ports in hotspots	Number of aggregated polygons
500	2241	1748	78.00%	438
800	1932	1468	75.98%	442
1000	1757	1353	77.01%	465
1200	1617	1261	77.98%	450
1500	1481	1157	78.12%	435

Tools

Get Citation

Copy Citation Text

Zeming Li, Liang Cheng, Daming Zhu, Zhaojin Yan, Chen Ji, Zhixin Duan, Min Jing, Ning Li, Shengkun Dongye, Yanruo Song, Jiahui Liu. Deep Learning and Spatial Analysis Based Port Detection[J]. Laser & Optoelectronics Progress, 2021, 58(20): 2028002

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites