摘要
为了能够较好地分割出舰船目标,实现后续的目标检测和识别,提出一种基于显著性检测的红外舰船图像分割方法。首先利用AC算法和FT算法对图像进行处理,将两种方法处理后的显著图合并,提高目标轮廓亮度,然后利用选大恒虚警检测的原理对图像进行二值化分割。通过对比其他几种分割算法,证明此算法有更强的抗干扰能力,分割效果更好。
To segment the ship target better and realize the subsequent target detection and recognition,an infrared ship image segmentation method based on saliency detection is proposed.Firstly,the AC and FT saliency detection algorithm are used to process the image,and the saliency images processed by the two methods are combined to improve the brightness of the target contour.Then the binary segmentation of the image is carried out based on the principle of Greatest of Constant False Alarm Rate Detection.Compared with the results of other segmentation algorithms,this new algorithm is proved to have stronger anti-interference ability and better segmentation effect.
作者
焦方浩
王中训
董云龙
JIAO Fanghao;WANG Zhongxun;DONG Yunlong(School of Physics and Electronic Information,Yantai University,Yantai 264005,China;Information Fusion Institute,Naval Aviation University,Yantai 264001,China)
出处
《烟台大学学报（自然科学与工程版）》
CAS
2023年第3期365-370,共6页
Journal of Yantai University(Natural Science and Engineering Edition)
基金
国家自然科学基金资助项目(61871392,62101583)。
关键词
红外舰船目标
显著性检测
恒虚警检测
图像分割
infrared ship target
saliency detection
Constant False Alarm Rate Detection
image segmentation
