生物芯片图像微阵列偏转角度计算及样点分割算法

Computation for deflection angle of microarray signal and image segmentation of biochip

本文首先介绍了样点分割在生物芯片技术中所处的重要意义,并分析了生物芯片样点偏转角度误差来源以及构成原因,提出了全自动化的图像微阵列偏转角度计算及样点分割的统一解决方案。算法依据生物芯片样点阵列空间分布特点,并基于重构原理,提出了环形投影方法和功率谱估计相结合的功率切片方法,并以此建立起样点阵列的像素位置、偏转角以及功率值的三维分布关系,实现斑点阵列中偏转角度的精确计算。在获得偏转角度值的条件下,采用定向投影法和功率谱估计方法,计算样点阵列的行列间距;利用行、列间距,在一维灰度和序列中实现分段搜索,获取光斑中心粗定位;利用轮廓图像投影后出现双次峰的现象,估算样点直径;最后,依据重心法调整斑点中心,同时采用邻域搜索算法调整斑点直径,最终实现生物芯片扫描图像的精确定位和分割。

For scan of biochip, the reasons of the deflection angle error are analyzed. Then automated method for calculating deflection angle of microarray signal and image segmentation is proposed. According to the spots distributing characteristics and reconstruction theory, power spectrum slice technique is presented, by which the 3-D distribution of space domain, deflection angle and power domain is constructed. So the deflection angle of microarray can be precisely calculated. Combined with fixed direction Fourier transform method and power spectrum transform method, column distance and row distance can be computed, which is used to search the spots center in the lists. In view of the fact that the outline image shows two sub-peaks if it is projected, spot diameter can be estimated. The spots center is adjusted based on gravity method, and associated with adjusting the diameter of spot by adjacent domain search method, and microarray image can be precisely located and segmented.