论文部分内容阅读
本文分析了甲襞毛细血管图像的特性，据此提出了甲襞毛细血管的软件自动跟踪窗口技术：即在时序图像上以视频速度搜索定位甲襞毛细血管，然后实施采样存贮。在计算机及图像输入设备给定的情况下，采样及存贮毛细血管图像窗口（其尺寸恒定）所需时间恒定，因而为了以视频速度跟踪甲襞毛细血管就必须将搜寻毛细血管所花时间降至最小。采用了两条措施来降低搜寻毛细血管的时间。第一条措施是，建立了甲襞毛细血管自动跟踪的数学模型，进而建立了ｆ函数，这种模型易于检测及搜索。第二条措施是，设计了最佳搜索算法，保证以最短时间搜索到毛细血管。最终实现了甲襞毛细血管的自动跟踪。
This paper analyzes the characteristics of capillaries in formazan. Based on this, we propose a software automatic tracking window technique for capillaries: search and locate capillaries at the video speed on time series images, and then implement sampling and storage. Given the constant time required to sample and store the capillary image window (which is of constant size) given the computer and image input device, in order to track the capillaries at the video speed, the time taken to search for the capillaries must be reduced To the minimum. Two steps have been taken to reduce the time spent in the search for capillaries. The first measure is to establish a mathematic model of automatic tracking of capillaries in capillaries, and then establish the f-function, which is easy to detect and search. The second measure is to design the best search algorithm, to ensure that the shortest time to search for capillaries. Finally realized the automatic tracking of capillaries.