基于改进DPA的空间小目标检测算法  

Space Small Targets Detection Based on Improved DPA

作　　者：张玉叶[1] 王春歆[2] 

机构地区：[1]海军航空工程学院青岛分院,山东青岛266000 [2]海军北海舰队指挥所,山东青岛266000

出　　处：《电子学报》2010年第3期556-560,共5页Acta Electronica Sinica

基　　金：国家863高技术研究发展计划(No.2006AA703213D)

摘　　要：针对以往动态规划算法(Dynamic Programming Algorithm,DPA)单点能量累积效率低、不能有效检测中途入场目标的问题,从三方面对DPA改进并应用于空间小目标检测.第一,对递归方程进行修改,利用多点累积方法克服工程应用中人为增加光学散焦导致目标信息丢失的问题;第二,将各个速度平面分别计算修改为速度更新,减少计算量;第三,增加对入场目标的特殊处理,克服不能有效检测中途入场目标的问题.利用实摄序列图像对比实验,证明改进的有效性.Former DPA（Dynamic Programming Algorithm）has problems of low efficiency in energy accumulation with single point and could not detect entrance target in time.Focusing on the problems we improved the DPA in three aspects and applied it in space small target detection.First,the recursive equation was improved to avoid information loss,which caused by manual defocus in the project,by multi-point accumulation.Second,the calculation of multi-velocity plane was simplified as velocity update to reduce the computational complexity.Third,special process was used to solve the problem of entrance targets.The contrastive experiment with real space images proved the validity of our improvements.

关 键 词：DPA 小目标检测 多点累积 入场目标 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]