基于微波光子学的激光夜视图像自动分割算法  

Automatic segmentation of laser night vision image based on microwave photonics

作　　者：李红睿 韩琛晔 崔红伟 LI Hongrui;HAN Chenye;CUI Hongwei(Hebei Polytechnic Institute,Shijiazhuang Hebei 050091,China)

机构地区：[1]河北工程技术学院,石家庄050091

出　　处：《激光杂志》2020年第8期124-128,共5页Laser Journal

基　　金：国家重点研究计划项目(No.2017YFA0700300)。

摘　　要：目前,大部分激光夜视图像分割算法主要通过对图像的增强、融合等预处理实现分割,分割过程对噪声的抑制力较弱,且分割过于复杂,提出一种基于微波光子学的激光夜视图像自动分割算法,首先使用基于微波光子学的激光夜视图像高功率微波测量系统,获取激光夜视图像中光波信号,然后基于该信号采用基于水平集的激光夜视图像分割算法,将C-V模型的分割曲线设成初始曲线,用于获取激光夜视图像全局收敛性,再在激光夜视图像的后续分割中导入分割曲线的局部信息,提升激光夜视图像分割性能。实验结果验证了所提算法的有效性,且具有分割误差小、分割效率快以及显著的抗噪性优势,应用价值极高。At present,most of the laser night vision image segmentation algorithms are mainly based on image enhancement and fusion pretreatment.The noise suppression of the segmentation process is weak and the segmentation is too complex.A laser night vision image automatic segmentation algorithm based on microwave photonics is proposed.Firstly,the laser night vision image high power microwave measurement system based on microwave photonics is used to obtain excitation.Based on the light microwave signal in night vision image,a level-set-based laser night vision image segmentation algorithm is used to set the segmentation curve of C-V model as an initial curve to obtain global convergence of laser night vision image.The local information of segmentation curve is introduced into subsequent segmentation of laser night vision image to improve the performance of laser night vision image segmentation.The experimental results verify the effectiveness of proposed algorithm,which has the advantages of small segmentation error,fast segmentation efficiency and remarkable anti-noise,and has high application value.

关 键 词：微波光子学 激光 夜视 图像 自动分割 水平集 

分 类 号：TN391[电子电信—物理电子学]