基于改进YOLOV3的遥感图像舰船目标检测  

Ship target detection in remote sensing image based on improved YOLOV3

作　　者：何维娟 江涛[1] 王欣 HE Wei-juan;JIANG Tao;WANG Xing(School of Mathematics and Computer Science,Yunnan Minzu University,Kunming 650500,China)

机构地区：[1]云南民族大学数学与计算机科学学院,云南昆明650500

出　　处：《云南民族大学学报（自然科学版）》2022年第1期89-93,共5页Journal of Yunnan Minzu University:Natural Sciences Edition

基　　金：国家自然科学基金(61363022);云南民族大学研究生创新基金(SJXY-2021-002).

摘　　要：对于舰船遥感目标检测中存在精度低、速度慢等问题,以YOLOV3为框架对YOLOV3网络进行改进,从而使其更适用于检测遥感舰船目标,为了使检测精度更高,在原本3个不同尺度的卷积特征图与深度残差网络中相应尺度的特征图进行融合前提下,增加了第4个尺度104×104,有效地学习样本的特征.实验结果表明:改进的YOLOV3网络能有效提高小尺度目标遥感舰船的检测精度,比原始YOLOV3算法的精度提高了12.1个百分点,整体性能优于原始的YOLOV3算法.This study aimed to provide solutions to problems of ship target detection in remote sensing such as low precision and slow speedby using the framework of YOLOV3 to improve YOLOV3 network,making it more practical in remote sensing of ship target detection.In order to ensure the higher detection accuracy,,in the original convolution characteristics of three different scale figures and depth characteristics of the corresponding scale figures in the residual network integration,A fourth scale 104×104 was added to effectively obtain the characteristics of the samples.The experimental results show that the improved YOLOV3 network can effectively improve the detection accuracy of small-scale target remote sensing ship,which is 12.1 percentage points higher than the original YOLOV3 algorithm,and the overall performance is better than the original YOLOV3 algorithm.

关 键 词：YOLOV3 目标检测 遥感舰船 特征图 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置]