首页> 外国专利> METHOD OF ARTIFICAL INTELLIGENCE TARGET LEARNING AND TARGET IDENTIFICATION FOR NEXT GENERATION NAVAL SHIP USING DIGITAL TWIN

METHOD OF ARTIFICAL INTELLIGENCE TARGET LEARNING AND TARGET IDENTIFICATION FOR NEXT GENERATION NAVAL SHIP USING DIGITAL TWIN

机译:基于数字孪生的下一代舰船人工智能目标学习与目标识别方法

摘要

Disclosed is an artificial intelligence target learning and target identification method for a next-generation trap using a digital twin. Receiving a target target image by the target target image input module; Performing a 3D rendering by applying a simulated maritime environment to the target target image received from the target target image input module by the target target 3D rendering module; The target target 3D rendering module configures outputting a 3D target image on which 3D rendering has been performed. According to the artificial intelligence target learning and target identification method for the next-generation ship using the digital twin described above, several chapters of the target target are configured to perform 3D rendering by applying various simulated maritime environments to the image, thereby constructing a database for the 3D target image In addition, there is an effect that can easily secure thousands of images under various marine environments required for AI-based recognition of the target trap. In addition, by performing AI-based learning using these 3D target images and through this, it is configured to accurately recognize the target of the sensor system image under AI-based, so that the enemy's target is automatically recognized and the commander's response can be determined more quickly. There is.
机译:公开了一种使用数字孪生的下一代陷阱的人工智能目标学习和目标识别方法。目标目标图像输入模块接收目标目标图像;通过将模拟海洋环境应用于目标目标3D渲染模块从目标目标图像输入模块接收的目标目标图像来执行3D渲染;目标目标3D渲染模块配置为输出已经执行3D渲染的3D目标图像。根据如上所述的使用数字双胞胎的下一代船的人工智能目标学习和目标识别方法,目标目标的几章被配置为通过将各种模拟海洋环境应用于图像来执行3D渲染,从而构建数据库用于3D目标图像的效果此外,在基于AI的目标陷阱识别所需的各种海洋环境下,可以轻松获得数千张图像的效果。此外,通过使用这些3D目标图像执行基于AI的学习,并以此来配置,以在基于AI的情况下准确识别传感器系统图像的目标,从而可以自动识别敌人的目标,并且指挥官的反应可以下定决心更快。有。

著录项

  • 公开/公告号KR102127657B1

    专利类型

  • 公开/公告日2020-06-29

    原文格式PDF

  • 申请/专利权人 HANWHA SYSTEMS CO. LTD.;

    申请/专利号KR20200050005

  • 发明设计人 김범준;

    申请日2020-04-24

  • 分类号G06T15;B63G13;F41J5;

  • 国家 KR

  • 入库时间 2022-08-21 11:04:21

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

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