...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>IEEE / ASME Transactions on Mechatronics >Wide-Angle Foveation for All-Purpose Use
【24h】

Wide-Angle Foveation for All-Purpose Use

机译：多功能用途的广角转向

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

This paper proposes a model of a wide-angle space-variant image that provides a guide for designing a fovea sensor. First, an advanced wide-angle foveated (AdWAF) model is formulated, taking all-purpose use into account. This proposed model uses both Cartesian (linear) coordinates and logarithmic coordinates in both planar projection and spherical projection. Thus, this model divides its wide-angle field of view into four areas, such that it can represent an image by various types of lenses, flexibly. The first simulation compares with other lens models, in terms of image height and resolution. The result shows that the AdWAF model can reduce image data by 13.5%, compared to a log-polar lens model, both having the same resolution in the central field of view. The AdWAF image is remapped from an actual input image by the prototype fovea lens, a wide-angle foveated (WAF) lens, using the proposed model. The second simulation compares with other foveation models used for the existing log-polar chip and vision system. The third simulation estimates a scale-invariant property by comparing with the existing fovea lens and the log-polar lens. The AdWAF model gives its planar logarithmic part a complete scale-invariant property, while the fovea lens has 7.6% error at most in its spherical logarithmic part. The fourth simulation computes optical flow in order to examine the unidirectional property when the fovea sensor by the AdWAF model moves, compared to the pinhole camera. The result obtained by using a concept of a virtual cylindrical screen indicates that the proposed model has advantages in terms of computation and application of the optical flow when the fovea sensor moves forward.
机译：本文提出了一个广角空间变异图像模型，为中央凹传感器的设计提供了指导。首先，在考虑了通用用途的前提下，制定了先进的广角凹形（AdWAF）模型。该提出的模型在平面投影和球面投影中都使用了笛卡尔坐标（线性）和对数坐标。因此，该模型将其广角视野分为四个区域，从而可以灵活地用各种类型的镜头表示图像。第一次仿真在图像高度和分辨率方面与其他镜头模型进行了比较。结果表明，与双极透镜模型相比，AdWAF模型可以将图像数据减少13.5％，两者在中心视场中的分辨率相同。使用提出的模型，通过原型中央凹透镜（广角中央凹（WAF）透镜）从实际输入图像重新映射AdWAF图像。第二个仿真与现有对数极性芯片和视觉系统使用的其他偏心模型进行了比较。第三模拟通过与现有的中央凹透镜和对数极透镜进行比较来估计尺度不变性。 AdWAF模型赋予其平面对数部分完整的尺度不变性，而中央凹透镜的球形对数部分最多具有7.6％的误差。与针孔相机相比，第四模拟计算光流，以便检查AdWAF模型的中央凹传感器移动时的单向性。通过使用虚拟圆柱屏幕的概念获得的结果表明，当中央凹传感器向前移动时，所提出的模型在光流的计算和应用方面具有优势。

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