基于多微面光纤面板的仿生复眼图像拼接及定位算法

为研究仿生复眼系统的图像拼接以及目标定位技术，使用三维设计制图软件SolidWorks对9个光学透镜组成的仿生复眼系统镜头支架进行设计；使用尺度不变特征变换算法对特征点进行提取，通过采用欧氏距离法和随机抽样一致性算法对特征点对进行筛选；采用加权平均法对图像进行融合；根据子眼镜头之间相互交叠的信息，选用了双目测距原理对空间目标进行三维定位。搭建了一套视场角为74.3°的视觉系统，得到了一副大视场的仿生复眼拼接图像，对空间4个不同位置的目标进行了定位，证明此仿生复眼系统能够获取大视场的拼接图像，能够对空间目标进行定位。

Image Mosaic and Positioning Algorithms of Bionic Compound Eye Based on Fiber Faceplate

In order to study the image mosaic and target location technology of bionic compound eye system, SolidWorks is used to design a lens holder for a bionic compound eye system composed of nine optical lenses. Feature points are extracted using scale-invariant feature transform algorithm. According to the overlapping information between the sub-eye lenses, the binocular distance measurement principle is adopted to locate spatial target in 3D. A vision system with a field of view of 74.3°is built, and a large field-of-view bionic compound eye mosaic image is obtained to locate the target in 4 different positions in space. It is proved that the bionic compound eye system can obtain large field-of-view mosaic images, and can locate the spatial target.