首页> 外文会议>Photonics Applications for Aviation, Aerospace, Commercial, and Harsh Environments V >Real-time atmospheric imaging and processing with hybrid adaptive optics and hardware accelerated lucky-region fusion (LRF) algorithm
【24h】

Real-time atmospheric imaging and processing with hybrid adaptive optics and hardware accelerated lucky-region fusion (LRF) algorithm

机译:利用混合自适应光学和硬件加速的幸运区域融合(LRF)算法进行实时大气成像和处理

获取原文
获取原文并翻译 | 示例

摘要

Atmospheric turbulences can significantly deteriorate the performance of long-range conventional imaging systems and create difficulties for target identification and recognition. Our in-house developed adaptive optics (AO) system, which contains high-performance deformable mirrors (DMs) and the fast stochastic parallel gradient decent (SPGD) control mechanism, allows effective compensation of such turbulence-induced wavefront aberrations and result in significant improvement on the image quality. In addition, we developed advanced digital synthetic imaging and processing technique, "lucky-region" fusion (LRF), to mitigate the image degradation over large field-of-view (FOV). The LRF algorithm extracts sharp regions from each image obtained from a series of short exposure frames and fuses them into a final improved image. We further implemented such algorithm into a VIRTEX-7 field programmable gate array (FPGA) and achieved real-time video processing. Experiments were performed by combining both AO and hardware implemented LRF processing technique over a near-horizontal 2.3km atmospheric propagation path. Our approach can also generate a universal real-time imaging and processing system with a general camera link input, a user controller interface, and a DVI video output.
机译:大气湍流会大大降低远程常规成像系统的性能,并给目标识别和识别带来困难。我们内部开发的自适应光学(AO)系统包含高性能可变形反射镜(DM)和快速随机平行梯度像样(SPGD)控制机制,可以有效补偿这种湍流引起的波前像差,从而带来显着改善在图像质量上。此外,我们开发了先进的数字合成成像和处理技术,即“幸运区”融合(LRF),以减轻大视场(FOV)上的图像退化。 LRF算法从从一系列短曝光帧中获得的每个图像中提取出锐利区域,并将它们融合为最终的改进图像。我们将这种算法进一步实现到VIRTEX-7现场可编程门阵列(FPGA)中,并实现了实时视频处理。通过将AO和硬件实现的LRF处理技术相结合,在近水平的2.3 km大气传播路径上进行了实验。我们的方法还可以生成具有通用摄像机链接输入,用户控制器接口和DVI视频输出的通用实时成像和处理系统。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号