• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>Targets and Backgrounds XII: Characterization and Representation >Information Processes in Visual and Object Buffers of Scene Understanding System for Reliable Target Detection, Separation from Background, and Identification
【24h】

Information Processes in Visual and Object Buffers of Scene Understanding System for Reliable Target Detection, Separation from Background, and Identification

机译：场景理解系统的视觉和对象缓冲区中的信息过程，用于可靠地进行目标检测，背景分离和识别

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Modem target recognition systems suffer from the lack of human-like abilities to understand the visual scene, detect, unambiguously identify and recognize objects. As result, the target recognition systems become dysfunctional if target doesn't demonstrate remarkably distinctive and contrast features that allow for unambiguous separation from background and identification upon such features. This is somewhat similar to visual systems of primitive animals like frogs, which can separate and recognize only moving objects. However, human vision unambiguously separates any object from its background. Human vision combines a rough but wide peripheral, and narrow but precise foveal systems with visual intelligence that utilize both scene and object contexts and resolve ambiguity and uncertainty in the visual information. Perceptual grouping is one of the most important processes in human vision, and it binds visual information into meaningful patterns and structures. Unlike the traditional computer vision models, biologically-inspired Network-Symbolic models convert image information into an "understandable" Network-Symbolic format, which is similar to relational knowledge models. The equivalent of interaction between peripheral and foveal systems in the network-symbolic system is achieved via interaction between Visual and Object Buffers and the top-level system of Visual Intelligence. This interaction provides recursive rough context identification of regions of interest in the visual scene and their analysis in the object buffer for precise and unambiguous separation of the object from background/clutter with following recognition of the target.
机译：现代的目标识别系统由于缺乏类似于人类的能力而无法理解视觉场景，检测，明确识别和识别物体。结果，如果目标物没有表现出明显的独特性和对比特征，从而无法与背景明确区分并根据这些特征进行识别，那么目标物识别系统就会失灵。这有点类似于像青蛙这样的原始动物的视觉系统，它们只能分离并识别运动的物体。但是，人类的视觉明确地将任何物体与其背景分离。人类视觉将粗糙但宽阔的外围，狭窄但精确的中央凹系统与视觉智能相结合，这些视觉智能利用场景和对象的上下文，并解决了视觉信息中的歧义和不确定性。感知分组是人类视觉中最重要的过程之一，它将视觉信息绑定为有意义的模式和结构。与传统的计算机视觉模型不同，受生物启发的网络符号模型将图像信息转换为“可理解的”网络符号格式，类似于关系知识模型。网络符号系统中外围系统和中央凹系统之间的等效交互是通过可视缓冲区和对象缓冲区与顶级可视智能系统之间的交互来实现的。这种交互提供了对视觉场景中感兴趣区域的递归粗略上下文识别，并在对象缓冲区中对其进行了分析，以在目标识别之后将对象与背景/杂波精确而明确地分离。

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