旋转不变的三维物体识别

Three-dimensional Object Recognition with Rotation-invariant by Structure Light

提出一种具有旋转不变性的三维物体识别的新方法。该方法通过结构照明的方法,使物体的高度分布以变形条纹的形式编码于二维强度像中。由于条纹图像包含有物体的高度分布信息,因而对条纹图的相关识别具有本征三维识别的特点。旋转不变性是通过使用多通道滤波器实现的,此滤波器可以由不同方向三维物体对应的变形条纹图像经计算机处理得到。相关识别方法可以用光学匹配空间滤波器实现。计算机模拟实验证明了这种方法的有效性,它不仅可以实现旋转不变的三维物体识别,还可以给出物体旋转角度的估计值。

An automatic method for rotation-invariant three-dimensional (3-D) object recognition was proposed. The method is based on the use of 3-D information contained in the deformed fringe pattern obtained when a grating is projected onto an object's surface. The deformed fringe patterns contain the height information about the objects, so the method has the characteristic of intrinsic 3-D recognition. The rotation invariance is achieved by using the multi-channel filter. The filter is composed of deformed fringe patterns of objects with different rotated angle. The proposed method was optically implemented by means of matching correlato (MC). The efficiency of rotation-invariant recognition of the method is verified by the computer simulation. And this method can estimate the rotate angle of the objects.