改进了以往针对二位图像的ＢＦＸＹ算法，将回溯算法应用到多灰度图像的多阈值分割边界自动跟踪中，抑制了噪声，提高了图像质量．根据超声成像（ＣＴ）体数据界面突出的特点，通过抽取切片内的边界，堆叠成三维表面进行显示，避免了贴片重构表面难于解决的分支问题．三维显示的结果图像清晰，缺陷分布一目了然．这一方法对我们可视理解声ＣＴ三维数据很有帮助． In the past, the BFXY algorithm for two-bit image was improved. The backtracking algorithm was applied to the automatic tracking of multi-threshold segmentation boundaries with multi-grayscale images, which suppressed the noise and improved the image quality. According to the characteristic of the interface of ultrasound imaging (CT) data, we can display the three-dimensional surface by extracting the inner boundary of the slice and avoiding the branch which is difficult to solve. Three-dimensional display of the results of a clear image, the distribution of defects at a glance. This method is helpful for us to understand the three-dimensional CT data.