基于水平集算法的污水中微生物图像分割研究

Segmentation research of the microbial images in the sewage based on level-set algorithm

日益严重的水污染现状促使着大批污水处理厂的出现,同时也使活性污泥法得到飞速的发展,在活性污泥法处理污水工艺中微生物的种类、数量以及其所处的生长阶段是污泥沉降性能的决定因素.主要研究水平集及其改进方法的细菌图像分割,通过镜检活性污泥中的微生物,然后对污水处理过程中的细菌图像进行分割识别,细菌图像分割类似于人眼对客观世界中不同对象进行分类的过程,它从图像中把相关的结构(或感兴趣区)分离出来,是细菌图像分析与识别首要解决的问题,而通过实验改进水平集分割算法对细菌图像分割,进而依据其分割结果判断微生物种类、数量,就可以预测污泥沉降性能,从而可以采取措施进一步改进工艺.

The current situation of increasingly serious water pollution not only promotes the emergence of a large number of sewage-treatment plants,but also makes the Activated Sludge Method(ASM)develop rapidly.The species,quantities and the growing stage of micro-organisms are the main factors that determine the performance of ASM for sludge sedimentation.This paper mainly studies the level set image segmentation and its improvement methods for bacteria images segmentation.The bacterial images during sewage-treatment processing are obtained with microscopic examination of microorganisms in activated sludge,and then segmented and identified with image segmentation methods.The processing of bacterial images segmentation is similar to the classification of objects with human eyes in real world.It separates relevant structures or areas of interest in the original images.How to conduct image segmentation is the primary problem to be solved in bacterial image analysis and identification.In this paper,by applying the proposed and improved level set segmentation algorithm,the bacterial images are segmented,with which the species and quantities of microorganisms can easily be retrieved,so as to judge the sludge sedimentation performance,which can be used as indicator for further improvement of ASM.