基于鲁棒背景估计的船舶显著性检测  

Salient Ship Detection Based on Robust Background Estimation

作　　者：姚婷婷[1] 张波 李鹏飞 柳晓鸣[1] Yao Tingting;Zhang Bo;Li Pengfei;Liu Xiaoming(School of Information Science and Technology,Dalian Maritime University,Dalian,Liaoning 116026,China)

机构地区：[1]大连海事大学信息科学技术学院,辽宁大连116026

出　　处：《激光与光电子学进展》2022年第8期121-129,共9页Laser & Optoelectronics Progress

基　　金：国家自然科学基金(62001078);中央高校基本科研业务费(3132020208)。

摘　　要：显著性检测技术可以快速有效地从海面背景中区分出前景船舶,因此基于显著性分析的船舶检测算法受到了广泛的研究关注。然而受到水面无规则背景噪声,如海浪、杂波、船舶尾迹等干扰,很难准确地获得船舶检测结果。针对这一问题,提出了一种基于鲁棒背景估计的船舶显著性检测算法。首先,对原始输入图像中的像素点进行聚类形成一系列超像素,并利用深度卷积网络求取每个超像素对应的特征描述。然后,为了有效抑制海面背景噪声对船舶检测性能的影响,构建了一种新的背景模板估计算法,并将其融入多尺度细胞自动机求解框架下,从而根据立体邻域空间中不同像素点的特征描述差异获得基于显著性分析的船舶检测结果。定性和定量实验结果表明,所提算法可以有效提高复杂背景下的船舶显著性检测效果。Foreground ships can be quickly and effectively detected from sea background with the help of salient detection technology.As a result,saliencyanalysisbased ship detection algorithms have received extensive research attention.However,obtaining accurate ship detection results influenced by irregular background noise,such as waves,clutter,and wakes,on the sea surface is challenging.A robust backgroundestimationbased salient ship detection algorithm has been proposed to solve the aforementioned problem.First,the input image is clustered into a set of superpixels,and the deep feature representation of each superpixel is extracted from a deep convolutional neural network.Then,a background noise estimation algorithm is proposed to effectively suppress the influence of background noise on ship detection and it is integrated into the solution framework of hierarchical cellular automata.Finally,the salient ship detection results can be obtained according to the difference in the feature description of various pixels in the stereo neighborhood space.Qualitative and quantitative experimental results demonstrate that the proposed algorithm could effectively enhance the salient ship detection effect under complex backgrounds.

关 键 词：图像处理 船舶显著性检测 背景估计 深度卷积网络 细胞自动机 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]