基于光学相干层析图像的珍珠质厚度自动测量算法

A method for automatic measurement of nacre thickness from optical coherence tomography images

利用OCT成像技术提供了一种珍珠厚度自动测量方法。该方法应用Canny边缘检测算子提取珍珠质的上边界,原始的边界利用先验知识拟合成圆,通过极坐标变换将图像从拟合的圆变换到极坐标系中,使用单侧差分与像素强度相结合的方法提取下边界。最后,用概率密度投影得到珍珠质厚度的概率分布。实验证明利用本方法进行厚度测量比原有手工测量结果的精度大幅提升,可以满足实际应用的需求。同时实验结果也证明了本方法对边界噪声、边界间断有很好的有效性与健壮性。

The method of automatic measurement of nacre thickness is from optical coherence tomo- graphy （OCT） images. It detects the upper boundary of nacre by Canny edge operator and the raw boundary is fitted to a circle by a priori. Then polar transform is applied to the image according to the fit- ted circle. At the following stage, the lower boundary is extracted by a transforming incorporating unilat- eral difference and intensity of pixels. Finaly, the probability distribution of nacre thickness is abtained by probability density projection. The proposed approach was evaluated using a large number of pearl optical coherence tomography images and achieved high accuracy, which could meet the requirements of practical applications. Besides, the method is robust to noise and boundary discontinuity. Experimental results demonstrate the effectiveness and robustness of the method.