同心拼图中基于狭缝图像段的近似深度矫正

Depth Correction in Concentric Mosaics: A Slit Image Segmentation Based Approximation

提出了一种对同心拼图进行近似深度矫正的新方法——基于狭缝图像段的矫正 .该方法通过对狭缝图像分段映射来对同心拼图进行矫正 ,合成畸变较小的结果图像 ;定义了狭缝图像的相似关系 ,通过两条相似狭缝图像之间的匹配 ,选择最佳的分段映射规则 ;并描述了为实现此目的而进行的场景采样方法 ,完成了从采样到矫正、绘制的整个流程 .与现有的其它同心拼图矫正方法相比 ,本方法具有适用范围更广 ,矫正效果更好的优点 .

The slit images used in concentric mosaics are first divided into segments, then the position and length of these segments are adjusted to correct the resulting new slit image. The concept of analogical slit images is presented, the segmentation and adjusting rule are extracted from two known analogical slit images. The complete process of implementing this approach is presented, including the sampling, slit image segmentation and mapping rule generation, rendering and correction. Experiments show the result of slit image approximation is better than the state-of-art depth correction methods.