高速运动目标的运动模糊图像复原研究

发布时间：2018-03-14 09:22

本文选题：运动模糊复原　切入点：点扩散函数可逆性　出处：《红外与激光工程》2017年04期 　论文类型：期刊论文

【摘要】：对单一图像进行运动模糊复原,存在模糊点扩散函数(PSF)难以估计以及图像反卷积的病态性问题。利用多个PSF具有联合可逆性的特点,针对运动目标观测,提出采用参数相同的多个成像设备共同对同一视场进行拍摄,来获取背景相同、曝光时间不同、目标模糊程度不同的观测图像;然后利用同一设备获取的序列图像进行目标的模糊PSF估计;并根据目标背景的运动模糊叠加特征,分别从观测图像中提取出完整的模糊目标图像;最后,对这些具有不同PSF的同一目标图像进行空间域迭代复原算式的联立求解。实验表明:该方法设计的目标获取装置对硬件条件要求较低,获取的图像更便于采用多点扩散函数联合进行图像复原,复原效果良好。
[Abstract]:For the motion blur restoration of a single image, it is difficult to estimate the fuzzy point diffusion function (PSF) and the ill-posed problem of image deconvolution. Using the characteristics of joint reversibility of multiple PSF, the observation of moving targets is studied. Several imaging devices with the same parameters are used to photograph the same field of view together to obtain the observation images with the same background, different exposure time and different target ambiguity. Then the fuzzy PSF estimation of the target is carried out by using the sequence images obtained by the same equipment; and the complete blurred target images are extracted from the observation images according to the motion fuzzy superposition characteristics of the target background. Finally, The same target images with different PSF are solved simultaneously by iterative algorithm in spatial domain. The experimental results show that the target acquisition device designed by this method requires less hardware conditions. The obtained image is more convenient for image restoration using multipoint diffusion function, and the restoration effect is good.
【作者单位】：海军航空工程学院青岛校区航空侦察教研室;北海舰队指挥所;
【基金】：山东省自然科学基金(ZR2013F0025)
【分类号】：TP391.41

【相似文献】