基于改进型YOLO算法的遥感图像舰船检测

Xikun Wang; Hongxu Jiang; Keyu Lin

doi:10.13700/j.bh.1001-5965.2019.0394

基于改进型YOLO算法的遥感图像舰船检测

Translated title of the contribution: Remote sensing image ship detection based on modified YOLO algorithm

Xikun Wang
, Hongxu Jiang^*
, Keyu Lin

^*Corresponding author for this work

School of Computer Science and Engineering

Beihang University

Research output: Contribution to journal › Article › peer-review

25 Scopus citations

Abstract

Although the target detection algorithm has achieved very good detection results in data sets such as PASCAL VOC.However, the accuracy of ship target detection in large-scale prediction images is very low.Therefore, according to the characteristics of the visible light reflection image, a feature mapping module is added on the basis of the YOLOv3-Tiny algorithm, which provides rich semantic information for the prediction layer.At the same time, a residual network is used in the feature extraction network, which improves the detection accuracy and effectively extracts ship features. Experimental results show that the detection accuracy of the optimized M-YOLO algorithm is 94.12%.Compared with the SSD and YOLOv3 algorithms, the detection accuracy of the M-YOLO algorithm is improved by 11.11% and 9.44%.

Translated title of the contribution	Remote sensing image ship detection based on modified YOLO algorithm
Original language	Chinese (Traditional)
Pages (from-to)	1184-1191
Number of pages	8
Journal	Beijing Hangkong Hangtian Daxue Xuebao/Journal of Beijing University of Aeronautics and Astronautics
Volume	46
Issue number	6
DOIs	https://doi.org/10.13700/j.bh.1001-5965.2019.0394
State	Published - 1 Jun 2020

Access to Document

10.13700/j.bh.1001-5965.2019.0394

Cite this

@article{33f79d26bc4b46bc9540b09fbd4bc974,

title = "基于改进型YOLO算法的遥感图像舰船检测",

abstract = "Although the target detection algorithm has achieved very good detection results in data sets such as PASCAL VOC.However, the accuracy of ship target detection in large-scale prediction images is very low.Therefore, according to the characteristics of the visible light reflection image, a feature mapping module is added on the basis of the YOLOv3-Tiny algorithm, which provides rich semantic information for the prediction layer.At the same time, a residual network is used in the feature extraction network, which improves the detection accuracy and effectively extracts ship features. Experimental results show that the detection accuracy of the optimized M-YOLO algorithm is 94.12\%.Compared with the SSD and YOLOv3 algorithms, the detection accuracy of the M-YOLO algorithm is improved by 11.11\% and 9.44\%.",

keywords = "Feature mapping module, Residual network, Ship detection, YOLOv3, YOLOv3-Tiny",

author = "Xikun Wang and Hongxu Jiang and Keyu Lin",

year = "2020",

month = jun,

day = "1",

doi = "10.13700/j.bh.1001-5965.2019.0394",

language = "繁体中文",

volume = "46",

pages = "1184--1191",

journal = "Beijing Hangkong Hangtian Daxue Xuebao/Journal of Beijing University of Aeronautics and Astronautics",

issn = "1001-5965",

publisher = "Beijing University of Aeronautics and Astronautics (BUAA)",

number = "6",

}

TY - JOUR

T1 - 基于改进型YOLO算法的遥感图像舰船检测

AU - Wang, Xikun

AU - Jiang, Hongxu

AU - Lin, Keyu

PY - 2020/6/1

Y1 - 2020/6/1

N2 - Although the target detection algorithm has achieved very good detection results in data sets such as PASCAL VOC.However, the accuracy of ship target detection in large-scale prediction images is very low.Therefore, according to the characteristics of the visible light reflection image, a feature mapping module is added on the basis of the YOLOv3-Tiny algorithm, which provides rich semantic information for the prediction layer.At the same time, a residual network is used in the feature extraction network, which improves the detection accuracy and effectively extracts ship features. Experimental results show that the detection accuracy of the optimized M-YOLO algorithm is 94.12%.Compared with the SSD and YOLOv3 algorithms, the detection accuracy of the M-YOLO algorithm is improved by 11.11% and 9.44%.

AB - Although the target detection algorithm has achieved very good detection results in data sets such as PASCAL VOC.However, the accuracy of ship target detection in large-scale prediction images is very low.Therefore, according to the characteristics of the visible light reflection image, a feature mapping module is added on the basis of the YOLOv3-Tiny algorithm, which provides rich semantic information for the prediction layer.At the same time, a residual network is used in the feature extraction network, which improves the detection accuracy and effectively extracts ship features. Experimental results show that the detection accuracy of the optimized M-YOLO algorithm is 94.12%.Compared with the SSD and YOLOv3 algorithms, the detection accuracy of the M-YOLO algorithm is improved by 11.11% and 9.44%.

KW - Feature mapping module

KW - Residual network

KW - Ship detection

KW - YOLOv3

KW - YOLOv3-Tiny

UR - https://www.scopus.com/pages/publications/85087381899

U2 - 10.13700/j.bh.1001-5965.2019.0394

DO - 10.13700/j.bh.1001-5965.2019.0394

M3 - 文章

AN - SCOPUS:85087381899

SN - 1001-5965

VL - 46

SP - 1184

EP - 1191

JO - Beijing Hangkong Hangtian Daxue Xuebao/Journal of Beijing University of Aeronautics and Astronautics

JF - Beijing Hangkong Hangtian Daxue Xuebao/Journal of Beijing University of Aeronautics and Astronautics

IS - 6

ER -

基于改进型YOLO算法的遥感图像舰船检测

Abstract

Access to Document

Other files and links

Fingerprint

Cite this