数字散斑时间序列相关三维面形测量方法

Shape Measurement by Digital Speckle Temporal Sequence Correlation Method

提出了一种测量三维面形的新方法。将随机数字散斑投影到参考平面上 ,深度方向等间距平移参考平面 ,用CCD摄取时间序列散斑参考集R(t)。然后用被测物体取代参考平面 ,获取物体调制散斑图像O。O中的任意子图像O(x ,y)与参考平面簇中对应位置的时间序列子图像R(x ,y ;t)之间的交叉相关值曲线呈高斯分布 ,其峰值位置就是被测物点的高度。该方法摆脱了以前数字散斑测量法水平相关思路 ,是真正的时间序列相关方法 ;且原理简单 ,测量精度高 ,不需要复杂的相位展开和校准过程 ,特别适用于测量突变面形和空间离散面形。根据测量结果的误差特征 ,提出了相关值加权邻域平均插值算法 ,能得到比邻域平均算法更好的测量结果。

A new method of three-dimension shape measurement was presented. The random digital speckles are projected on the reference plane, which is moved by equal distance in the direction of depth. Temporal sequence speckle images are acquired with CCD as reference set R (t). Replace the reference plane by object, the speckle image O modulated by the object is also grasped with CCD. The correlation value between subimage O (x, y) and the corresponding subimages R (x, y; t) is of approximately quadratic curve distribution. The position of the curve peak is the height value of the object point. The method goes out of the limit of original horizontal correlation in previous method and is a true temporal sequence correlation method. It has the advantages of simple principle and high measurement precision without complicated phase unwrapping and calibration, and can be used to measure the discontinuous and spatial isolation shape. According to the characteristics of error, a new interpolation method based on correlation weight was also presented.